Defects in SFS 2.0 Which Affect the Working-Set

by Stephen Gold, SFS vice-chair

FINAL DRAFT, last updated 19 July 2001

1.0 Executive Summary

Significant defects have recently been discovered in the SFS 2.0 benchmark suite, causing it to be withdrawn. Each of these defects tends to reduce the size of the benchmark's working-set as the number of load-generating processes is reduced. Since test-sponsors have great freedom in deciding how many processes to use, the resulting variations in the working-set hamper meaningful comparison of SFS 2.0 test results.

Defect #1 manifests when the requested ops/sec/process is in the hundreds. In observed incidents, the working-set of the benchmark may be reduced as much as 27%.

Defect #2 manifests whenever the requested ops/sec/process is 26 or more. The working-set of the benchmark may be reduced by a factor of three or more.

Defect #3 manifests only when the requested ops/sec/process is 500 or more. The working-set of the benchmark may be be reduced two orders of magnitude or more.

Defect #4: Even if the first three defects were corrected, the cache profile of the working-set would still vary with the number of load-generating processes used. As processes are reduced, the fileset distribution narrows.

Because of these defects, many of the published SFS 2.0 results are not comparable. Comparability exists only for results that were run on servers which would have cached the intended working-set of the benchmark. Based on simulation, it is believed that 115 out of the 248 approved SFS 2.0 results meet this criterion for comparability.

2.0 Technical Background

The SFS 2.0 benchmark suite was released by SPEC December 1997. SFS 2.0 is used to evaluate and compare the performance of fileservers under steady NFS loads. SFS 2.0 was withdrawn by the SPEC Open Systems Group in June 2001 due to the defects which are the subject of this document.

Important features of SFS 2.0 included:

• coordination of multiple load-generating clients
• coordination of multiple load-generating processes on each client
• portability between different load-generating clients
• a dataset which grows in direct proportion to the requested load
• adaptive regulation of the load generated by each process
• measurement of both achieved throughput and average response-time
• a 300-second warmup phase before each measurement phase
• run-rules intended to create a level playing-field for competition
• reporting rules intended to provide all information needed to reproduce each published result

The SFS 2.0 suite consists of two workloads, SPECsfs97.v2 and SPECsfs97.v3. The differences between the two workloads boil down to the operation-mixes used. Data-set generation, file selection, and rate regulation are done exactly the same way for both workloads. Thus, defects in these areas affect both workloads.

Much of the remainder of this discussion assumes that the reader is a licensee with access to the source-code. But sufficient detail is presented to allow a non-licensee to grasp the broad outlines of how the benchmark works and what went wrong.

2.1 Working-Set

The working-set of an NFS workload over an interval is the number of distinct filesystem blocks accessed during the interval.

All modern NFS servers have some ability to cache frequently-accessed filesystem blocks in their buffer cache for later reuse. This is done to reduce the probability that any particular access will require a (costly) disk read.

On many servers, the buffer cache is smaller than the working-set of the SFS 2.0 benchmark over its 300-second measurement phase. (One reason for the 300-second warmup phase in SFS 2.0 is to warm up the cache sufficiently to approximate the steady-state behavior.) For such servers, changes to the working-set size can have a large impact on the number of disk operations, the utilization of the CPU and disks, achievable throughput, and average response-times.

2.2 How the SFS 2.0 dataset is Generated

The design intent of SFS 2.0 was that the dataset would be 10 MBytes per op/s of requested load.

Looking at the implementation (sfs_c_chd.c) we see that each load-generating process has its own dataset, rooted in its own subdirectory, with no sharing between processes. The dataset of each process consists of four so-called "working sets". Each "working set" is used for a different group of NFS operations:

Io_working_set

used for "I/O operations": GETATTR, SETATTR, LOOKUP, READ, WRITE, FSSTAT, ACCESS, and non-failing LOOKUP operations

Non_io_working_set

used for CREATE, REMOVE, and failing LOOKUP operations

Dir_working_set

used for READDIR and READDIRPLUS operations

Symlink_working_set

used for READLINK operations

In this paper we are mainly concerned with the Io_working_set.

The code that sizes the "working sets" is in sfs_c_man.c. Num_io_files gets set to roughly 390 times the requested op/sec for the process. The INIT phase ensures that this number of "I/O" files (with names like file_en.00001) get created on the server. Since the average file size is about 27 KBytes, this works out to about 10 MBytes per requested op/sec.

Note that the acheived ops/sec reported in the benchmark disclosure generally differs slightly from the requested ops/sec.

2.3 How "I/O" Files are Selected

The design intent of SFS 2.0 was that only 10% of the files would be accessed in the course of a run, and furthermore that the files would be accessed according to a Poisson distribution, causing some files to be accessed much more frequently than others.

Num_working_io_files is set to roughly 10% of Num_io_files. The init_fileinfo() function in sfs_c_chd.c randomly selects Num_working_io_files of the "I/O" files to go into the Io_working_set.

To facilitate generation of the Poisson distribution, the Io_working_set is divided into eight or more access-groups of roughly equal size. An access-group in the Io_working_set typically contains about 100 files. The number of access-groups is always a multiple of four. To be precise:

   group_cnt = 8 + ((Num_working_io_files/500) * 4)

Each access-group is assigned a relative weight, called "next_value". The sum of the weights from 0 to i is stored in Io_working_set.entries[i].range. Then the sum of all the weights is stored in Io_working_set.max_range.

When a load-generating process wants to generate an NFS operation, it calls the do_op() function in sfs_c_chd.c, which calls op(). If the operation chosen by do_op() is one which uses the Io_working_set, op() calls randfh() with file_type=Sfs_io_file. Some operations (like READ and WRITE) may call randfh() more than once, until they find a file with particular properties.

In order to pick an access-group, randfh() generates a random value between 0 and (Io_working_set.max_range - 1), inclusive. It then does a binary search, starting at Io_working_set.max_range/2 - 1 and proceeding until it finds a group such that either:

• Io_working_set.entries[group - 1].range < value && value <= Io_working_set.entries[group].range

or

• group == 0 && value <= Io_working_set.entries[group].range

If the random value were uniformly distributed and the ".range" fields were monotonically non-decreasing, the probability of selecting a particular group would be:

   (Io_working_set.entries[group].range - Io_working_set.entries[group - 1].range)/Io_working_set.max_range  (if group > 1)

or 

   Io_working_set.entries[0].range/Io_working_set.max_range  (if group = 0)

which would equal next_value/Io_working_set.max_range . (Recall that a different next_value is computed for each group.)

In order to pick a file, randfh() generates random offsets within the selected group. If the operation is a READ or a non-append WRITE, randfh() keeps generating offsets until it finds a file that is large enough for the current operation.

2.4 How Request Rates are Regulated

Between calls to do_op(), the load-generating processes sleeps by calling msec_sleep(). Each load-generating process regulates its request rate independantly from the other processes. This is done by periodically adjusting the average sleep-time, Target_sleep_mspc. The goal of the adjustment is to find a value of Target_sleep_mspc which causes the process to generate its share of the requested total rate, which is Child_call_load.

The same adjustment algorithm is used for both the warmup phase and the measurement phase, just with different parameters. It is fairly simple. A process starts the warmup phase with Target_sleep_mspc calculated based on sleeping 50% the time. After a period of time has elapsed, the process calls check_call_rate() to adjust Target_sleep_mspc.

The check_call_rate() function begins by comparing actual requests generated since start of the current phase (Reqs_this_test) with a target based on multiplying the Child_call_load by elapsed time in the current phase (elapsed_time). The difference between the target and actual requests is added to the normal request count for the upcoming check-period (req_target_per_period) and that is the number of requests to be attempted in the upcoming check-period. In other words, the process attempts to be completely caught up by the end of the upcoming check-period.

The check_call_rate() function sets Target_sleep_mspc to reflect the number of requests to be attempted, the duration of a check-period, and the expected work time per request in the upcoming check-period. Since the actual work times are not known, the code substitutes (as an approximation) the average work time for all requests so far in the current phase.

During the warmup phase, each check-period lasts for two seconds. Since the warmup phase is 300 seconds, a total of 150 checks are made during the warmup phase. By that time the Target_sleep_mspc has hopefully converged to a stable value. During the measurement phase (which lasts for another 300 seconds) the checking and adjustments continue, but the check-period is increased to 10 seconds.

The load-generating process does not attempt to sleep for exactly Target_sleep_msec each time. Instead, it generates a random number of milliseconds (rand_sleep_msec) which is distributed uniformly between roughly 50% and 150% of Target_sleep_msec. This randomized sleep interval is passed to msec_sleep(), which in turn calls the library function select(). The timeout argument passed to select is based directly on rand_sleep_msec.

3.0 Description of the Defects

3.1 Defect #1: request-rate regulation is unstable, causing processes to sleep for many seconds, aka "The oscillation defect"

discovered May 7, 2001

The rate adjustment algorithm described in Section 2.3 is highly aggressive. For instance, if the process generates twice the requested rate during first check-period of the warmup phase, it will try to recover completely during the second check-period. To do so, it will attempt to make zero requests during that period. This is rather extreme considering that there are still 149 periods remaining during the warmup phase!

Because the adjustment algorithm is so aggressive, it tends to be unstable, particularly when the process is asked to generate a large number of requests per second.

The instability of the adjustment algorithm is exacerbated by difficulties in accurately controlling the interval between requests. On many UNIX systems, the select() timeout resolution is one clock-tick, which typically means 10 msec. For instance, msec_sleep(1) is likely to sleep for at least 10 msec, for an error of >900% with respect to the requested interval!

If the adjustment algorithm sees that the process is falling behind with Target_sleep_mspc=1, it is likely to try Target_sleep_mspc=0. This causes a sudden jump in the request-rate, lasting for an entire period. This can put the process so far ahead that it will stall for the period after that.

Furthermore, SFS 2.0 tests are usually run with more load-generating processes than CPUs. This means that processes whose select() timeout has expired may have to wait for another process to relinquish the CPU before it can return from select() and generate another request. This increases the upredictability of the msec_sleep() function and tends to introduce further instability.

The instability, when present, can be observed by enabling the CHILD_XPOINT debugging code in check_call_rate(), which can be accomplished by setting DEBUG=6 in the sfs_rc file. If you see debug output containing the string " 0 call, rqnd " then you will know that one of the load-generating processes generated zero calls for that check-period.

The zero-call behavior has been observed both during warmup phases and measurement phases. With respect to measurement phases, the behavior has been observed for request-rates ranging from 200 to 400 ops/sec/process, affecting 6-27% of the adjustment periods. It is possible that the defect could occur for request-rates outside-this range.

When a load-generating process experiences a zero-call period during the measurement phase, its portion of the fileset receives no accesses for ten seconds. Over that interval, the working-set of the benchmark is reduced by 1/N from the intended working-set, where N is the total number of load-generating processes.

Smaller oscillations in the request-rate, say from 50% to 150% of the requested rate, would also tend to reduce the working-set, though the effect is harder to quantify.

To summarize:

When the requested ops/sec/process is in the hundreds, instability in the rate regulation algorithm can cause the request-rate of a process to oscillate wildly. Sometimes the request-rate of a process goes to zero for ten seconds at a time. The frequency and severity of the problem are poorly understood. In observed incidents, the working-set of the benchmark was reduced by 6%-27%.

3.2 Defect #2: I/O access-group probabilities can round to zero, aka "The distribution defect"

discovered May 7, 2001

The next_value variable (i.e., the weight given to access-group i) is calculated as:

   next_value = (int) (1000 * (cumulative_lambda_power /
                     (e_to_the_lambda * cumulative_x_factorial)))

where:

• lambda = group_cnt / 2
• cumulative_lambda_power is lambda to the ith power

and

• cumulative_x_factorial is (i+1)!.

Whenever the requested ops/sec/process is in the range 26 to 38, there will be 16 access groups, lambda = 8, and e_to_the_lambda = 2980.96. This causes access-group #0 to have next_value of:

      (int) (1000 * (1.0 / (2980.96 * 1.0)))) = (int)0.335 = 0

So Io_working_set.entries[0].range == 0, which mean that group #0, containing roughly 6% of the Io_working_set files is never selected by randfh(Sfs_io_file). In effect, the probability of access, which should have been 0.03%, got rounded down to zero.

As the request-rates increase, so do group_cnt, lambda, and e_to_the_lambda. next_value=0 occurs for more and more access-groups, making them inaccessible. This phenomenon starts at the margins of the distribution (groups 0 and group_cnt-1, where the Poisson probabilities are lowest) and spreads inward toward the mode, as illustrated in the following table:

Requested    Number of I/O access-groups:       Inaccessible
ops/proc     total  inaccessible  accessible    I/O access-groups:
  1-12          8     0  (0%)          8         none
 13-25         12     0  (0%)         12         none
 26-38         16     1  (6%)         15         {0}
 39-51         20     2 (10%)         18         {0-1}
 52-64         24     3 (13%)         21         {0-2}
 65-76         28     5 (18%)         23         {0-3, 27}
 77-89         32     8 (25%)         24         {0-5, 30-31}
 90-102        36    11 (31%)         25         {0-6, 32-35}
103-115        40    13 (33%)         27         {0-7, 35-39}
116-128        44    17 (39%)         27         {0-9, 37-43}
129-141        48    19 (40%)         29         {0-10, 40-47}
142-153        52    22 (42%)         30         {0-11, 42-51}
 ...
193-205        68    35 (51%)         33         {0-18, 52-67}
 ...
295-307       100    60 (60%)         40         {0-30, 71-99}
 ...
398-410       132    87 (66%)         45         {0-44, 90-131}
411-423       136    90 (66%)         46         {0-45, 92-135}
424-435       140    93 (66%)         47         {0-47, 95-139}
436-448       144    97 (67%)         47         {0-49, 97-143}
 ...
475-487       156   107 (69%)         49         {0-54, 104-155}
488-499       160   111 (69%)         49         {0-56, 106-159}

The "inaccessible" groups in this table are just the ones which have next_value=0.

Here is the same data presented in graphical form:

Note that below 26 requested ops/sec/proc, all the access-groups are accessible and this defect has no effect.

The trend would continue past 500 requested ops/sec/proc, except that at that point the existence of defect #3 complicates the issue.

Inaccessible I/O access-groups reduce the number of files accessed from what the benchmark intended. Of the "Files accessed for I/O operations" printed in the client log, it might be the case that only 1/3 were truly accessible via those operations. Thus the I/O operations get concentrated over a smaller set of files than was intended.

The inaccessible files are precisely the ones that would be accessed least frequently in a correct Poission distribution. On the other hand, these files are also the ones that are most likely to miss in the server's buffer cache, so their inaccessibility could have a relatively large impact on buffer-cache miss-rates. Buffer-cache miss-rates could affect the average response-times reported by the benchmark as well as the peak throughputs.

Don Capps of Hewlett-Packard has created a simulation of the SFS 2.0 benchmark which can quickly determine, for a given SFS 2.0 result, what fraction of the Io_working_set access-groups were inaccessible due to this defect.

To summarize:

Whenever the requested ops/sec/process is 26 or more, there are some I/O access-groups which cannot be accessed because their probability of access is zero. The number of inaccessible access-groups generally increases as the request-rate increases. At 475 requested ops/sec/process, the working-set of the "I/O operations" in SFS 2.0 is reduced by more than a factor of three.

3.3 Defect #3: I/O access-group ranges can be negative, aka "The floating-point overflow defect"

discovered May 7, 2001

Recall that the next_value (the weight of access-group i) is calculated as:

   next_value = (int) (1000 * (cumulative_lambda_power /
                     (e_to_the_lambda * cumulative_x_factorial)))

All three variables in the right-hand side of the assignment are double-precision. If the load-generating client implements IEEE Standard 754 floating point (as most do) the largest accurately-represented value is roughly 2e+308.

For 500 requested ops/sec/process, there are 164 groups and lambda=82. Something strange happens for group #162, since cumulative_lambda_power (82 to the 162nd power) is roughly 1e+310, which is represented as Infinity. The denominator is also Infinity, since e_to_the_lambda (4.094e+35) times cumulative_x_factorial (1.22969e+289) also overflows 2e+308. So the quotient is Infinity divided by Infinity, or NaN (not-a-number). 1000 times NaN is still NaN. And converting NaN to an int results in (2^31 - 1) or 2147483647.

Now, the "range" values for each access-group are declared as ints. It so happens that previous_range = Io_working_set.entries[161].range = 970.

Adding 2147483647 and 970 using 32-bit twos-complement arithmetic (which is what most compilers generate for ints) yields -2147482679, which is stored in Io_working_set.entries[162].range.

Since the NaN also occurs for i=163, next_value is (2^31 - 1) again. Io_working_set.entries[163].range and Io_working_set.max_range both get set to 968.

Now one of the assumptions of the binary search algorithm (namely, that the "range" fields are monotonically non-decreasing) has been violated for the last two access-groups. This violation is the basis for defect #3.

For 164 groups this defect is not a serious matter, since the binary search starts at group=81 and never reaches the last two access-groups for any random value between 0 and 967.

As the request-rate increases beyond 512 requested ops/sec/process, so do group_cnt, lambda, and e_to_the_lambda. next_value=NaN occurs for more and more access-groups. As long as the NaNs are confined the highest access-groups and there are an even number of NaNs, this defect has little or no effect.

However, if the number of NaNs happens to be odd instead of even (as happens when there are 168 or 176 groups) then Io_working_set.max_range will be a large negative integer. In this case the random value generated will be a positive number in the range 0 to 1-max_range. The effect is to cause the vast majority of accesses to go to one or two access-groups.

For example, consider a process with 168 access-groups. There are seven NaNs and max_range = -2147482689. The binary search algorithm picks a value in the range 0 to 2147482690. The binary search algorithm starts at group=83, which has Io_working_set.entries[83].range = 469.

By far the most likely scenario is that the random value is greater than 966, in which case the binary search examines groups 125, 146, 157, 162, 165, and 166, before settling on group=167. Over the course of a 300-second measurement phase (in which the process generates on the order of a million I/O operations) the expected number of random values between 0 and 966 is less than 1. In effect, the working-set for I/O operations has been reduced to a single access-group.

For exactly 304 access-groups, max_range is zero, causing a division by zero error which terminates the benchmark with a core dump.

When the number of access-groups exceeds 304, the NaNs invade the middle access-groups, where the binary search starts. Even if the number of NaNs is even, the binary search algorithm can get seriously confused by them. Once again, the effect is to cause the vast majority of accesses to go to one or two access-groups.

For example, consider a process with 324 access-groups. This time there are 184 NaNs and max_range=-162. The binary search algorithm picks a value in the range 0 to 161. The binary search algorithm starts at group=161. But:

   Io_working_set.entries[161].range = 0 and
   Io_working_set.entries[162].range = 2147483647.

By walking through the code, you can see that group #161 is selected for value=0 and group #162 is selected for 0 < value < 162:

        if (work_set->entries[group].range == value)
            break;
        if (work_set->entries[group].range > value) {
            ...
        } else if (work_set->entries[group].range < value) {
            if (work_set->entries[group+1].range > value) {
                group++;
                break;           

Thus, out of 324 groups, only two groups are accessible, and group #162 gets over 99% of the I/O accesses.

Don Capp's simulator calculates the group-access probabilities, generates random numbers, and performs the binary search using algorithms equivalent to those in the SFS 2.0 load generator. The number of accesses simulated is always 300 times the requested ops/sec. This is slightly unrealistic because (for various reasons) the actual number of I/O accesses per requested op is not really unity. Nevertheless the simulator provides a very convincing illustration of how the number of I/O access-groups actually accessed varies for different numbers of requested ops/sec/process.

The following table, generated using the simulator, shows how the number of groups with next_value=0 and next_value=NaN varies for selected numbers of access-groups. The last column shows how many access-groups were actually selected by the binary search algorithm after it had been invoked millions of times.

Requested    number of I/O access-groups:
ops/proc     total  next_value=0  next_value=NaN  selected via binary search
488-499       160      111              0              49 (31%)
500-512       164      112              2              49 (30%)
513-525       168      111              7               2  (1%) 
526-538       172      109             12              46 (27%)
539-551       176      108             17               2  (1%)
552-564       180      106             22              45 (25%)
 ...
590-602       192      103             36              44 (23%)
 ...
629-641       204       99             50              44 (22%)
 ...
667-679       216       96             56              30 (14%)
680-692       220       96             68              30 (14%)
693-705       224       94             73               2 (<1%)
706-717       228       93             78              32 (14%)
718-730       232       92             82              33 (14%)
 ...
757-769       244       95             96              32 (13%)
770-782       248       96            100              32 (13%)
 ...
795-807       256      100            109               3  (1%)
 ...
885-897       284      113            140              28 (10%)
898-910       288      113            145               1 (<1%)
911-923       292      116            149               1 (<1%)  
924-935       296      118            153               1 (<1%)
936-948       300      120            158               1 (<1%)
949-961       304      122            162              [core-dump]
962-974       308      122            167               1 (<1%)
975-987       312      125            171               1 (<1%)
988-999       316      127            175               1 (<1%)
1000-1012     320      129            180               2 (<1%)
1013-1025     324      131            184               2 (<1%)
1026-1038     328      133            188               1 (<1%)

Here is the same data presented in graphical form:

To summarize:

NaNs are generated whenever the requested ops/sec/process is 500 or more. These NaNs cause violation of the assumptions underlying the algorithm used to select access-groups.

When the number of NaNs is even and less than 150, the number of groups accessed generally declines as the request-rate increases, reaching 28 groups at 897 requested ops/sec/process. The working-set of the "I/O operations" in SFS 2.0 is reduced by up to an order of magnitude.

When the number of NaNs is odd and/or greater than 150, the number of groups accessed is usually three or less, with most of the accesses going to a single group. The working-set of the "I/O operations" in SFS 2.0 is reduced by at least two orders of magnitude.

3.4 Defect #4: fileset distribution narrows as processes are reduced

discovered June 5, 2001

Defects #2 and #3 can be remedied by making minor changes to the software which sets up the ranges. But even if the Poisson distribution were implemented as intended, there would still be a serious problem with the benchmark. The problem lies in the way the Poisson parameter "lambda" varies with the per-process request rate.

Although buffer cache algorithms vary in detail from server to server, it is reasonable to assume that the cacheability of an access-group increases with its frequency of access. So when analyzing cache behavior, it makes sense to sort the access-groups in order of probability, so that the rank of the most frequently-accessed group is 1 and the rank of the least frequently-accessed group is N. Plotting a group's probability of access against its sorted rank produces a "cache profile" of the workload.

As the number of groups grows, the cache profile becomes more and more concentrated on the left edge of the graph. In other words, a greater and greater fraction of the accesses occur in the busiest parts of the fileset.

For a simple illustration, imagine a server which always hits in the most-frequently accessed 1/8 of the groups and always misses in the remaining 7/8 of the groups. For a target load is 50,000 ops/sec, the Io_working_set size is always roughly 50 GBytes. By simply varying the number of load-generating processes, the fraction of accesses which are to cached groups can be varied from 20% to 84%, as shown below:

Procs   Ops/proc  Groups/proc    Cached_Groups Lambda    % cached
5000       10         8            5000           4        20.6%
1500       33        16            3000           8        28.1%
 625       80        32            2500          16        38.5%
 263      190        64            2104          32        52.1%
 128      390       128            2048          64        68.3%
  62      806       256            1859         128        84.2%

The only constraints on the number of load-generating processes come from the performance capabilities of the server-under-test and compliance with the Uniform Access Rule (UAR).

Of course, if the server-under-test caches the benchmark's working-set over the entire measurement phase, then the changes in the cache profile should have little or no impact on the results.

To summarize:

Even if the other three defects were corrected, the cacheability of the working-set would still vary with the number of load-generating processes used. This variability would invalidate any comparison of SFS 2.0 results for servers that were tested with different number of processes, unless it could be shown that both servers were operating entirely out of cache.

4.0 Effect on Approved Results

A total of 248 SFS 2.0 results have been approved by SPEC. 247 of those results have been published on the SPEC website. (The remaining result is awaiting the vendor's decision whether or not to publish.)

Don Capps's simulator has been used to demonstrate that 115 of the published SFS 2.0 results would have fit in the servers' memory even if the intended working-set had been achieved. Since the defects described in this report affect only the working-set of the benchmark, such results can be considered comparable to one another. Summarized simulation results for all 248 approved SFS 2.0 results are presented in Appendix A.

Of the 248 approved results, the requested ops/sec/process at peak ranges from 18.89 to 1000. Only one result exceeded the 499 requested ops/sec/process threshold for triggering defect #3. 230 results (93%) exceeded the 25 requested ops/sec/process threshold for triggering defect #2. Because defect #1 has not been adequately modeled, it is unknown how many of the approved results might have triggered that defect.

Of all the approved SFS 2.0 results, the five with the highest requested ops/sec/process at the peak were:

1. 4054 SPECsfs97.v2 ops per second with an overall response time of 0.94 ms: 1000 requested ops/sec/process (not published yet)
2. 14186 SPECsfs97.v3 ops per second with an overall response time of 1.55 ms: 447 requested ops/sec/process
3. 10011 SPECsfs97.v3 ops per second with an overall response time of 1.24 ms: 420 requested ops/sec/process
4. 13056 SPECsfs97.v3 ops per second with an overall response time of 1.46 ms: 406.25 requested ops/sec/process
5. 17286 SPECsfs97.v3 ops per second with an overall response time of 1.33 ms: 402.5 requested ops/sec/process

All five of these results were submitted in 2001. Aside from these five results, no other approved results exceeded 400 requested ops/sec/process.

Using the tables in this report, one can see that:

• Result 1 was affected by defect #3 at the peak, to the extent that only 2 of each process's 320 access-groups were actually accessed. In other words, the set of groups accessed was reduced by a factor of 160.
• Results 2 through 5 were affected by defect #2 at the peak:
  • Result 2 was affected to the extent that only 47 each process's 144 access-groups were actually accessed. In other words, the set of groups accessed was reduced by a factor of 3.06x.
  • Result 3 was affected to the extent that only 46 each process's 136 access-groups were actually accessed. In other words, the set of groups accessed was reduced by a factor of 3.09x.
  • Results 4 and 5 were affected to the extent that only 45 each process's 132 access-groups were actually accessed. In other words, the set of groups accessed was reduced by a factor of 2.93x.

According to the simulator, none of these five results was obtained with a server that could operate entirely out of cache with the intended working-set. So in each case the performance of the server-under-test was exaggerated by the defects described in this report.

Many other results were affected as well. See Appendix A for more information about specific results.

5.0 Conclusions

Substantial defects have recently been discovered in the SFS 2.0 benchmark suite, causing it to be withdrawn. Each of these defects tends to reduce the size of the benchmark's working-set as the number of load-generating processes is reduced. Since test-sponsors have great freedom in deciding how many processes to use, the resulting variations in the working-set hamper meaningful comparison of SFS 2.0 test results for servers that do not cache the entire dataset.

Defect #1 manifests when the requested ops/sec/process is in the hundreds. Instability in the rate regulation algorithm can cause the request-rate of a process to go to zero for ten seconds at a time. The frequency and severity of the problem are poorly understood. In observed incidents, the working-set of the benchmark was reduced by 6%-27%.

Defect #2 manifests whenever the requested ops/sec/process is 26 or more. Some I/O access-groups become inaccessible because their probability of access is rounded down to zero. The number of inaccessible access-groups generally increases as the request-rate increases. At 475 requested ops/sec/process, the working-set of the "I/O operations" in SFS 2.0 is reduced by more than a factor of three. About 92% of the approved SFS 2.0 runs suffered from inaccessible access-groups because of this defect, though in about half of those cases the reported performance is thought to have been unaffected because the server-under-test could hold the intended working-set in cache.

Defect #3 manifests only when the requested ops/sec/process is 500 or more. Some I/O access-groups become inaccessible because propagation of floating-point overflows into NaNs result in violation of an assumption underlying the access-group selection algorithm. When the number of NaNs is even and less than 150, the working-set of the "I/O operations" in SFS 2.0 is reduced by up to an order of magnitude. When the number of NaNs is odd and/or greater than 150, the defect becomes much more serious and the working-set of the "I/O operations" in SFS 2.0 is reduced at least two orders of magnitude. Only one approved SFS 2.0 result has been affected by this defect. As of the publication of this report, that result has not been published on the SPEC website.

Even if the first three defects were corrected, the cache profile of the working-set would still vary with the number of load-generating processes used. This variability would invalidate any comparison of SFS 2.0 results for servers that were tested with different number of processes, unless it could be shown that both servers were operating entirely out of cache.

Because of these defects, many of the published SFS 2.0 results are not comparable. Comparability exists only for results that were run on servers which would have cached the intended working-set of the benchmark. Based on simulation, it is believed that 115 out of the 248 approved SFS 2.0 results meet this criterion for comparability.

Appendix A: Simulation Results

Here are the summarized simulation results for all 248 approved SFS 2.0 results. There are a few caveats to be considered in interpreting these results:

• defect #1 is not simulated
• the simulator considers only the I/O working-set, not at any of the other working sets
• the simulator assumes that all RAM and NVRAM listed in the disclosure are available for buffer-cache
• the simulator assumes that there are exactly 300 accesses to the I/O working set for each requested op/sec

Here are the 98 approved SPECsfs97.v2 results, sorted by achieved throughput at peak:

Approved Result	Peak requested load, in ops/sec	Number of processes	Peak requested ops/sec/process	SUT memory (MBytes)	I/O files created (GBytes)	SFS 2.0 I/O access-groups	SFS 2.0 I/O access-groups accessed (percent)	SFS 2.0 est. I/O files accessed: cacheable+uncacheable	SFS 2.0 est. accesses to uncacheable I/O files	SFS 3.0 est. I/O files accessed: cacheable+uncacheable	SFS 3.0 est. accesses to uncacheable I/O files	Status of the result, with respect to defects
1146 SPECsfs97.v2 ops per second with an overall response time of 3.60 ms	1200	9	133.33	264	468000 (11)	432	261 (60%)	20790+4392	6165	20790+4392	83214	probably affected
1300 SPECsfs97.v2 ops per second with an overall response time of 3.27 ms	1398	9	155.33	264	545220 (13)	504	306 (57%)	20790+6318	12510	20790+6318	126261	probably affected
2010 SPECsfs97.v2 ops per second with an overall response time of 1.88 ms	2000	8	250.00	512	780000 (19)	672	296 (44%)	31968+0	0	31968+0	98056	probably affected
2217 SPECsfs97.v2 ops per second with an overall response time of 5.58 ms	2200	64	34.38	1028	858000 (21)	1024	768 (100%)	71744+0	0	71744+0	0	I/O working set was cacheable; probably not affected
2366 SPECsfs97.v2 ops per second with an overall response time of 4.14 ms	2400	39	61.54	264	936000 (23)	936	819 (88%)	20787+48282	288951	20787+48282	378534	probably affected
2691 SPECsfs97.v2 ops per second with an overall response time of 3.98 ms	2700	24	112.50	264	1053000 (26)	960	648 (75%)	20784+41016	283320	20784+41016	460296	probably affected
2903 SPECsfs97.v2 ops per second with an overall response time of 8.07 ms	2880	56	51.43	512	1123200 (28)	1344	1176 (88%)	40320+41552	161504	40320+41552	281456	probably affected
3141 SPECsfs97.v2 ops per second with an overall response time of 3.82 ms	3100	40	77.50	1088	1209000 (30)	1280	1080 (75%)	77120+0	0	77120+0	38800	probably affected
3372 SPECsfs97.v2 ops per second with an overall response time of 5.34 ms	3360	56	60.00	3072	1310400 (32)	1344	1176 (88%)	96432+0	0	96432+0	0	I/O working set was cacheable; probably not affected
3425 SPECsfs97.v2 ops per second with an overall response time of 7.56 ms	3400	48	70.83	1024	1326000 (33)	1344	1008 (88%)	80640+9504	9504	80640+9504	96192	probably affected
3827 SPECsfs97.v2 ops per second with an overall response time of 3.92 ms	3840	88	43.64	3072	1497600 (37)	1760	1936 (92%)	118448+0	0	118448+0	0	I/O working set was cacheable; probably not affected
3930 SPECsfs97.v2 ops per second with an overall response time of 3.11 ms	3996	66	60.55	544	1558440 (39)	1584	1386 (88%)	42834+72006	368478	42834+72006	532356	probably affected
4054 SPECsfs97.v2 ops per second with an overall response time of 0.94 ms (unpublished)	4000	4	1000.00	800	1560000 (39)	1280	20 (2%)	968+0	0	968+0	328628	probably affected
4201 SPECsfs97.v2 ops per second with an overall response time of 3.36 ms	4194	63	66.57	544	1635660 (40)	1764	1323 (88%)	42840+68229	372834	42840+68229	581427	probably affected
4217 SPECsfs97.v2 ops per second with an overall response time of 3.32 ms	4200	84	50.00	544	1638000 (41)	1680	1848 (92%)	42840+85512	470316	42840+85512	587160	probably affected
4508 SPECsfs97.v2 ops per second with an overall response time of 2.74 ms	4500	216	20.83	7168	1755000 (43)	2592	2592 (100%)	156168+0	0	156168+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
4549 SPECsfs97.v2 ops per second with an overall response time of 6.28 ms	4596	84	54.71	1024	1792440 (44)	2016	1764 (88%)	80640+50820	133056	80640+50820	305256	probably affected
4672 SPECsfs97.v2 ops per second with an overall response time of 7.61 ms	4698	72	65.25	4096	1832220 (45)	2016	1512 (88%)	123912+0	0	123912+0	0	I/O working set was cacheable; probably not affected
4752 SPECsfs97.v2 ops per second with an overall response time of 3.73 ms	4752	45	105.60	544	1853280 (46)	1800	1215 (75%)	42840+65520	387540	42840+65520	726345	probably affected
5023 SPECsfs97.v2 ops per second with an overall response time of 3.89 ms	4998	54	92.56	544	1949220 (48)	1944	1350 (69%)	42822+74682	481410	42822+74682	800010	probably affected
5095 SPECsfs97.v2 ops per second with an overall response time of 3.60 ms	5100	48	106.25	544	1989000 (49)	1920	1296 (75%)	42816+73152	472032	42816+73152	828000	probably affected
5240 SPECsfs97.v2 ops per second with an overall response time of 5.69 ms	5200	56	92.86	3072	2028000 (50)	2016	1400 (69%)	121856+0	0	121856+0	0	I/O working set was cacheable; probably not affected
5303 SPECsfs97.v2 ops per second with an overall response time of 4.64 ms	5250	84	62.50	1024	2047500 (51)	2016	1764 (88%)	80640+69720	234360	80640+69720	449904	probably affected
5402 SPECsfs97.v2 ops per second with an overall response time of 3.92 ms	5376	56	96.00	544	2096640 (52)	2016	1400 (69%)	42840+84896	581392	42840+84896	905520	probably affected
5550 SPECsfs97.v2 ops per second with an overall response time of 4.78 ms	5500	84	65.48	1024	2145000 (53)	2352	1764 (88%)	80640+65772	226800	80640+65772	505260	probably affected
5952 SPECsfs97.v2 ops per second with an overall response time of 3.96 ms	5929	132	44.92	3072	2312310 (57)	2640	2904 (92%)	182028+0	0	182028+0	0	I/O working set was cacheable; probably not affected
6030 SPECsfs97.v2 ops per second with an overall response time of 4.89 ms	6000	210	28.57	1216	2340000 (58)	3360	2520 (100%)	95760+100800	378210	95760+100800	466200	probably affected
6071 SPECsfs97.v2 ops per second with an overall response time of 3.41 ms	6048	102	59.29	1056	2358720 (59)	2448	2142 (88%)	83130+92820	355572	83130+92820	611592	probably affected
6155 SPECsfs97.v2 ops per second with an overall response time of 2.85 ms	6296	112	56.21	1056	2455440 (61)	2688	2352 (88%)	83104+97328	404656	83104+97328	674240	probably affected
7025 SPECsfs97.v2 ops per second with an overall response time of 8.25 ms	7050	72	97.92	4096	2749500 (68)	2592	1800 (69%)	167112+0	0	167112+0	0	I/O working set was cacheable; probably not affected
7046 SPECsfs97.v2 ops per second with an overall response time of 4.79 ms	7050	90	78.33	1216	2749500 (68)	2880	2430 (75%)	95760+79830	294390	95760+79830	713880	probably affected
7206 SPECsfs97.v2 ops per second with an overall response time of 2.76 ms	7200	128	56.25	1024	2808000 (70)	3072	2688 (88%)	80640+125568	621568	80640+125568	924032	probably affected
7431 SPECsfs97.v2 ops per second with an overall response time of 2.30 ms	7560	54	140.00	4960	2948400 (73)	2592	1566 (60%)	158112+0	0	158112+0	0	I/O working set was cacheable; probably not affected
7462 SPECsfs97.v2 ops per second with an overall response time of 3.53 ms	7440	66	112.73	1056	2901600 (72)	2640	1782 (75%)	83160+86790	394812	83160+86790	953502	probably affected
7612 SPECsfs97.v2 ops per second with an overall response time of 7.46 ms	7548	72	104.83	4096	2943720 (73)	2880	1944 (75%)	173232+0	0	173232+0	0	I/O working set was cacheable; probably not affected
7750 SPECsfs97.v2 ops per second with an overall response time of 2.67 ms	7800	48	162.50	1056	3042000 (76)	2688	1632 (57%)	83136+67968	298800	83136+67968	1049760	probably affected
8165 SPECsfs97.v2 ops per second with an overall response time of 3.04 ms	8280	576	14.38	14336	3229200 (80)	6912	6912 (100%)	285120+0	0	285120+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
8170 SPECsfs97.v2 ops per second with an overall response time of 2.72 ms	8100	128	63.28	1024	3159000 (79)	3072	2688 (88%)	80640+149504	828416	80640+149504	1146880	probably affected
8292 SPECsfs97.v2 ops per second with an overall response time of 6.03 ms	8280	168	49.29	3072	3229200 (80)	3360	3696 (92%)	241920+12600	12600	241920+12600	76104	probably affected
9079 SPECsfs97.v2 ops per second with an overall response time of 2.56 ms	9180	54	170.00	4960	3580200 (89)	3240	1728 (53%)	172800+0	0	172800+0	0	I/O working set was cacheable; probably not affected
9133 SPECsfs97.v2 ops per second with an overall response time of 3.60 ms	9096	72	126.33	5248	3547440 (88)	3168	2160 (62%)	195984+0	0	195984+0	0	I/O working set was cacheable; probably not affected
9825 SPECsfs97.v2 ops per second with an overall response time of 5.92 ms	9760	112	87.14	5120	3806400 (95)	3584	3024 (75%)	243488+0	0	243488+0	0	I/O working set was cacheable; probably not affected
10384 SPECsfs97.v2 ops per second with an overall response time of 3.93 ms	10350	198	52.27	4096	4036500 (101)	4752	4158 (88%)	295218+0	0	295218+0	46728	probably affected
10715 SPECsfs97.v2 ops per second with an overall response time of 5.43 ms	10593	117	90.54	8192	4131270 (103)	4212	2925 (69%)	250146+0	0	250146+0	0	I/O working set was cacheable; probably not affected
10724 SPECsfs97.v2 ops per second with an overall response time of 3.09 ms	10992	72	152.67	9344	4286880 (107)	3744	2160 (62%)	222912+0	0	222912+0	0	I/O working set was cacheable; probably not affected
11382 SPECsfs97.v2 ops per second with an overall response time of 2.99 ms	11400	192	59.38	2112	4446000 (111)	4608	4032 (88%)	166272+164928	580224	166272+164928	1053888	probably affected
11806 SPECsfs97.v2 ops per second with an overall response time of 5.16 ms	11700	240	48.75	3072	4563000 (114)	4800	5280 (92%)	241920+120240	248160	241920+120240	531840	probably affected
11823 SPECsfs97.v2 ops per second with an overall response time of 5.62 ms	11760	168	70.00	5120	4586400 (114)	4704	3528 (88%)	313992+0	0	313992+0	6552	probably affected
13435 SPECsfs97.v2 ops per second with an overall response time of 3.17 ms	13496	128	105.44	2112	5263440 (131)	5120	3456 (75%)	166272+141952	545664	166272+141952	1553536	probably affected
13605 SPECsfs97.v2 ops per second with an overall response time of 3.15 ms	13520	1536	8.80	14336	5272800 (132)	12288	16896 (92%)	494592+0	0	494592+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
14081 SPECsfs97.v2 ops per second with an overall response time of 1.99 ms	13992	48	291.50	10240	5456880 (136)	4608	1872 (41%)	208704+0	0	208704+0	0	I/O working set was cacheable; probably not affected
14140 SPECsfs97.v2 ops per second with an overall response time of 2.91 ms	14300	320	44.69	2048	5577000 (139)	6400	7040 (92%)	161280+280000	1384960	161280+280000	1802240	probably affected
14279 SPECsfs97.v2 ops per second with an overall response time of 6.28 ms	14400	168	85.71	5120	5616000 (140)	5376	4536 (75%)	357000+0	0	357000+0	163128	probably affected
14941 SPECsfs97.v2 ops per second with an overall response time of 5.64 ms	15120	224	67.50	6144	5896800 (147)	6272	4704 (88%)	405440+0	0	405440+0	45024	probably affected
15053 SPECsfs97.v2 ops per second with an overall response time of 5.76 ms	15030	567	26.51	3648	5861700 (146)	9072	6804 (100%)	286902+205821	610659	286902+205821	824418	probably affected
15235 SPECsfs97.v2 ops per second with an overall response time of 1.54 ms	15200	50	304.00	3200	5928000 (148)	5000	2000 (42%)	221000+0	0	221000+0	1136350	probably affected
15270 SPECsfs97.v2 ops per second with an overall response time of 1.91 ms	15984	48	333.00	10240	6233760 (156)	5184	1968 (38%)	225888+0	0	225888+0	0	I/O working set was cacheable; probably not affected
15421 SPECsfs97.v2 ops per second with an overall response time of 4.20 ms	15400	264	58.33	6144	6006000 (150)	6336	5544 (88%)	438240+0	0	438240+0	63360	probably affected
16060 SPECsfs97.v2 ops per second with an overall response time of 6.13 ms	16524	486	34.00	3648	6444360 (161)	7776	5832 (100%)	287226+250290	861678	287226+250290	1128492	probably affected
16138 SPECsfs97.v2 ops per second with an overall response time of 3.00 ms	16000	256	62.50	2048	6240000 (156)	6144	5376 (88%)	161280+296960	1614592	161280+296960	2240768	probably affected
16832 SPECsfs97.v2 ops per second with an overall response time of 2.98 ms	16698	216	77.31	16384	6512220 (163)	6912	5832 (75%)	416448+0	0	416448+0	0	I/O working set was cacheable; probably not affected
17437 SPECsfs97.v2 ops per second with an overall response time of 4.50 ms	17500	450	38.89	3648	6825000 (170)	9000	9900 (92%)	287100+245700	790650	287100+245700	1333800	probably affected
17674 SPECsfs97.v2 ops per second with an overall response time of 5.69 ms	17600	224	78.57	12288	6864000 (171)	7168	6048 (75%)	437024+0	0	437024+0	0	I/O working set was cacheable; probably not affected
18235 SPECsfs97.v2 ops per second with an overall response time of 3.63 ms	18240	480	38.00	8192	7113600 (178)	7680	5760 (100%)	595200+0	0	595200+0	0	I/O working set was cacheable; probably not affected
18293 SPECsfs97.v2 ops per second with an overall response time of 3.80 ms	19000	216	87.96	3456	7410000 (185)	6912	5832 (75%)	272160+202176	676944	272160+202176	1794960	probably affected
18431 SPECsfs97.v2 ops per second with an overall response time of 4.33 ms	18700	264	70.83	8192	7293000 (182)	7392	5544 (88%)	495792+0	0	495792+0	11352	probably affected
18860 SPECsfs97.v2 ops per second with an overall response time of 3.57 ms	18400	1536	11.98	20480	7176000 (179)	12288	16896 (92%)	681984+0	0	681984+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
19755 SPECsfs97.v2 ops per second with an overall response time of 3.77 ms	20394	243	83.93	3456	7953660 (199)	7776	6561 (75%)	272160+232308	888165	272160+232308	2091258	probably affected
20176 SPECsfs97.v2 ops per second with an overall response time of 5.58 ms	21120	224	94.29	12288	8236800 (206)	8064	5600 (69%)	501536+0	0	501536+0	0	I/O working set was cacheable; probably not affected
20406 SPECsfs97.v2 ops per second with an overall response time of 6.77 ms	20250	720	28.12	6144	7897500 (197)	11520	8640 (100%)	483840+184320	339120	483840+184320	545760	probably affected
20460 SPECsfs97.v2 ops per second with an overall response time of 3.78 ms	20400	480	42.50	8192	7956000 (199)	9600	10560 (92%)	622560+0	0	622560+0	84960	probably affected
20683 SPECsfs97.v2 ops per second with an overall response time of 3.21 ms	21000	420	50.00	3072	8190000 (205)	8400	9240 (92%)	241920+399840	2002140	241920+399840	2619120	probably affected
20925 SPECsfs97.v2 ops per second with an overall response time of 4.18 ms	21010	264	79.58	12288	8193900 (205)	8448	7128 (75%)	518496+0	0	518496+0	0	I/O working set was cacheable; probably not affected
22017 SPECsfs97.v2 ops per second with an overall response time of 4.04 ms	22000	450	48.89	3648	8580000 (214)	9000	9900 (92%)	287100+391950	1651050	287100+391950	2331000	probably affected
22885 SPECsfs97.v2 ops per second with an overall response time of 5.17 ms	24300	252	96.43	12288	9477000 (237)	9072	6300 (69%)	574812+0	0	574812+0	0	I/O working set was cacheable; probably not affected
23687 SPECsfs97.v2 ops per second with an overall response time of 3.32 ms	23496	360	65.27	3072	9163440 (229)	10080	7560 (88%)	241920+377640	2069640	241920+377640	3260880	probably affected
23815 SPECsfs97.v2 ops per second with an overall response time of 4.39 ms	24288	264	92.00	12288	9472320 (237)	9504	6600 (69%)	571032+0	0	571032+0	0	I/O working set was cacheable; probably not affected
25639 SPECsfs97.v2 ops per second with an overall response time of 9.94 ms	25440	720	35.33	8192	9921600 (248)	11520	8640 (100%)	645120+191520	317520	645120+191520	535680	probably affected
25849 SPECsfs97.v2 ops per second with an overall response time of 3.79 ms	25776	480	53.70	12288	10052640 (251)	11520	10080 (88%)	739200+0	0	739200+0	0	I/O working set was cacheable; probably not affected
25973 SPECsfs97.v2 ops per second with an overall response time of 1.55 ms	26700	80	333.75	6400	10413000 (260)	8960	3360 (39%)	368480+0	0	368480+0	1518960	probably affected
27097 SPECsfs97.v2 ops per second with an overall response time of 3.70 ms	27000	252	107.14	8192	10530000 (263)	10080	6804 (75%)	616392+0	0	616392+0	757512	probably affected
27210 SPECsfs97.v2 ops per second with an overall response time of 3.73 ms	27000	252	107.14	8192	10530000 (263)	10080	6804 (75%)	616392+0	0	616392+0	757512	probably affected
27646 SPECsfs97.v2 ops per second with an overall response time of 4.25 ms	27840	240	116.00	15360	10857600 (271)	10560	7200 (62%)	601680+0	0	601680+0	0	I/O working set was cacheable; probably not affected
27688 SPECsfs97.v2 ops per second with an overall response time of 3.46 ms	28496	512	55.66	4096	11113440 (278)	12288	10752 (88%)	322560+497152	2415104	322560+497152	3596288	probably affected
28265 SPECsfs97.v2 ops per second with an overall response time of 3.87 ms	28050	396	70.83	18432	10939500 (273)	11088	8316 (88%)	743688+0	0	743688+0	0	I/O working set was cacheable; probably not affected
30555 SPECsfs97.v2 ops per second with an overall response time of 1.56 ms	30400	100	304.00	6400	11856000 (296)	10000	4000 (42%)	442000+0	0	442000+0	2272700	probably affected
30684 SPECsfs97.v2 ops per second with an overall response time of 3.59 ms	30496	512	59.56	4096	11893440 (297)	12288	10752 (88%)	322560+560640	2870272	322560+560640	4072960	probably affected
30703 SPECsfs97.v2 ops per second with an overall response time of 3.78 ms	31200	2160	14.44	21504	12168000 (304)	25920	25920 (100%)	1069200+0	0	1069200+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
31803 SPECsfs97.v2 ops per second with an overall response time of 4.42 ms	31680	352	90.00	24576	12355200 (309)	12672	8800 (69%)	748000+0	0	748000+0	0	I/O working set was cacheable; probably not affected
40218 SPECsfs97.v2 ops per second with an overall response time of 6.31 ms	40000	400	100.00	16384	15600000 (390)	14400	10000 (69%)	950400+0	0	950400+0	112800	probably affected
40938 SPECsfs97.v2 ops per second with an overall response time of 3.03 ms	40800	4608	8.85	36864	15912000 (398)	36864	50688 (92%)	1502208+0	0	1502208+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
44593 SPECsfs97.v2 ops per second with an overall response time of 3.62 ms	46984	784	59.93	7168	18323760 (458)	18816	16464 (88%)	564480+781648	3590720	564480+781648	5526416	probably affected
46840 SPECsfs97.v2 ops per second with an overall response time of 2.95 ms	46800	6000	7.80	44032	18252000 (457)	48000	66000 (92%)	1758000+0	0	1758000+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
49718 SPECsfs97.v2 ops per second with an overall response time of 3.84 ms	49476	784	63.11	7168	19295640 (483)	18816	16464 (88%)	564480+845152	4165392	564480+845152	6203792	probably affected
91218 SPECsfs97.v2 ops per second with an overall response time of 1.59 ms	91200	288	316.67	19200	35568000 (890)	29952	12672 (41%)	1306944+0	0	1306944+0	6829056	probably affected
103966 SPECsfs97.v2 ops per second with an overall response time of 2.91 ms	104976	1008	104.14	262144	40940640 (1025)	40320	27216 (75%)	2388960+0	0	2388960+0	0	I/O working set was cacheable; probably not affected
182548 SPECsfs97.v2 ops per second with an overall response time of 1.59 ms	183600	576	318.75	38400	71604000 (1792)	59904	25344 (41%)	2634048+0	0	2634048+0	13928832	probably affected
250573 SPECsfs97.v2 ops per second with an overall response time of 1.58 ms	249984	1024	244.12	51200	97493760 (2441)	86016	37888 (44%)	4000768+0	0	4000768+0	19857408	probably affected

Here are the 150 approved SPECsfs97.v3 results, sorted by achieved throughput at peak:

Approved Result	Peak requested load, in ops/sec	Number of processes	Peak requested ops/sec/process	SUT memory (MBytes)	I/O files created (GBytes)	SFS 2.0 I/O access-groups	SFS 2.0 I/O access-groups accessed (percent)	SFS 2.0 est. I/O files accessed: cacheable+uncacheable	SFS 2.0 est. accesses to uncacheable I/O files	SFS 3.0 est. I/O files accessed: cacheable+uncacheable	SFS 3.0 est. accesses to uncacheable I/O files	Status of the result, with respect to defects
1101 SPECsfs97.v3 ops per second with an overall response time of 5.14 ms	1100	20	55.00	1028	429000 (10)	480	420 (88%)	31640+0	0	31640+0	0	I/O working set was cacheable; probably not affected
1612 SPECsfs97.v3 ops per second with an overall response time of 7.83 ms	1600	56	28.57	512	624000 (15)	896	672 (100%)	40320+12096	19208	40320+12096	32368	probably affected
1844 SPECsfs97.v3 ops per second with an overall response time of 8.74 ms	1840	56	32.86	512	717600 (17)	896	672 (100%)	40320+19544	46424	40320+19544	66360	probably affected
1928 SPECsfs97.v3 ops per second with an overall response time of 5.67 ms	1936	56	34.57	3072	755040 (18)	896	672 (100%)	62944+0	0	62944+0	0	I/O working set was cacheable; probably not affected
2141 SPECsfs97.v3 ops per second with an overall response time of 4.62 ms	2142	66	32.45	3072	835380 (20)	1056	792 (100%)	70026+0	0	70026+0	0	I/O working set was cacheable; probably not affected
2201 SPECsfs97.v3 ops per second with an overall response time of 4.98 ms	2200	32	68.75	1088	858000 (21)	896	672 (88%)	58848+0	0	58848+0	0	I/O working set was cacheable; probably not affected
2206 SPECsfs97.v3 ops per second with an overall response time of 2.56 ms	2200	10	220.00	2112	858000 (21)	760	350 (49%)	36190+0	0	36190+0	0	I/O working set was cacheable; probably not affected
2207 SPECsfs97.v3 ops per second with an overall response time of 5.59 ms	2208	56	39.43	3072	861120 (21)	1120	1232 (92%)	67144+0	0	67144+0	0	I/O working set was cacheable; probably not affected
2212 SPECsfs97.v3 ops per second with an overall response time of 2.92 ms	2200	12	183.33	2112	858000 (21)	768	396 (55%)	40056+0	0	40056+0	0	I/O working set was cacheable; probably not affected
2298 SPECsfs97.v3 ops per second with an overall response time of 4.51 ms	2319	21	110.43	264	904410 (22)	840	567 (75%)	20790+32277	193599	20790+32277	358134	probably affected
2364 SPECsfs97.v3 ops per second with an overall response time of 4.40 ms	2370	66	35.91	3072	924300 (23)	1056	792 (100%)	77484+0	0	77484+0	0	I/O working set was cacheable; probably not affected
2419 SPECsfs97.v3 ops per second with an overall response time of 3.76 ms	2400	108	22.22	5632	936000 (23)	1296	1296 (100%)	83484+0	0	83484+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
2461 SPECsfs97.v3 ops per second with an overall response time of 4.62 ms	2490	21	118.57	264	971100 (24)	924	630 (62%)	20790+33495	216972	20790+33495	403242	probably affected
2633 SPECsfs97.v3 ops per second with an overall response time of 6.54 ms	2640	56	47.14	1024	1029600 (25)	1120	1232 (92%)	80528+0	0	80528+0	15344	probably affected
2649 SPECsfs97.v3 ops per second with an overall response time of 3.34 ms	2640	144	18.33	5632	1029600 (25)	1728	1728 (100%)	91008+0	0	91008+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
2943 SPECsfs97.v3 ops per second with an overall response time of 6.39 ms	2960	56	52.86	1024	1154400 (28)	1344	1176 (88%)	80640+3472	3472	80640+3472	43904	probably affected
2944 SPECsfs97.v3 ops per second with an overall response time of 4.04 ms	3000	42	71.43	1024	1170000 (29)	1176	882 (88%)	80262+0	0	80262+0	47922	probably affected
2991 SPECsfs97.v3 ops per second with an overall response time of 3.66 ms	2997	36	83.25	544	1168830 (29)	1152	972 (75%)	42840+31788	106992	42840+31788	284616	probably affected
3121 SPECsfs97.v3 ops per second with an overall response time of 3.49 ms	3120	30	104.00	544	1216800 (30)	1200	810 (75%)	42840+28260	90240	42840+28260	315390	probably affected
3135 SPECsfs97.v3 ops per second with an overall response time of 4.13 ms	3150	42	75.00	1024	1228500 (30)	1176	882 (88%)	80640+3822	3822	80640+3822	63672	probably affected
3219 SPECsfs97.v3 ops per second with an overall response time of 5.76 ms	3200	56	57.14	3072	1248000 (31)	1344	1176 (88%)	92232+0	0	92232+0	0	I/O working set was cacheable; probably not affected
3279 SPECsfs97.v3 ops per second with an overall response time of 3.59 ms	3306	30	110.20	544	1289340 (32)	1200	810 (75%)	42840+32970	115830	42840+32970	360330	probably affected
3324 SPECsfs97.v3 ops per second with an overall response time of 3.55 ms	3336	32	104.25	544	1301040 (32)	1280	864 (75%)	42848+32992	120352	42848+32992	364896	probably affected
3345 SPECsfs97.v3 ops per second with an overall response time of 3.75 ms	3348	36	93.00	544	1305720 (32)	1296	900 (69%)	42840+35676	142992	42840+35676	370872	probably affected
3379 SPECsfs97.v3 ops per second with an overall response time of 2.00 ms	3498	12	291.50	544	1364220 (34)	1152	468 (41%)	42840+9336	16872	42840+9336	408396	probably affected
3491 SPECsfs97.v3 ops per second with an overall response time of 3.66 ms	3498	30	116.60	544	1364220 (34)	1320	900 (62%)	42840+32850	126780	42840+32850	405150	probably affected
3576 SPECsfs97.v3 ops per second with an overall response time of 4.79 ms	3560	88	40.45	3072	1388400 (34)	1760	1936 (92%)	107888+0	0	107888+0	0	I/O working set was cacheable; probably not affected
3608 SPECsfs97.v3 ops per second with an overall response time of 3.32 ms	3630	24	151.25	544	1415700 (35)	1248	720 (62%)	42840+30480	114432	42840+30480	442008	probably affected
3610 SPECsfs97.v3 ops per second with an overall response time of 5.97 ms	3628	128	28.34	3584	1414920 (35)	2048	1536 (100%)	119808+0	0	119808+0	0	I/O working set was cacheable; probably not affected
3648 SPECsfs97.v3 ops per second with an overall response time of 3.64 ms	3648	30	121.60	544	1422720 (35)	1320	900 (62%)	42840+36000	147480	42840+36000	443580	probably affected
3710 SPECsfs97.v3 ops per second with an overall response time of 2.83 ms	3748	20	187.40	320	1461720 (36)	1280	660 (55%)	25200+43120	345240	25200+43120	691280	probably affected
3713 SPECsfs97.v3 ops per second with an overall response time of 5.78 ms	3696	56	66.00	3072	1441440 (36)	1568	1176 (88%)	97832+0	0	97832+0	0	I/O working set was cacheable; probably not affected
3717 SPECsfs97.v3 ops per second with an overall response time of 2.85 ms	3748	20	187.40	320	1461720 (36)	1280	660 (55%)	25200+43120	345240	25200+43120	691280	probably affected
3946 SPECsfs97.v3 ops per second with an overall response time of 2.14 ms	4080	15	272.00	544	1591200 (39)	1380	630 (44%)	42840+19785	66060	42840+19785	556275	probably affected
3967 SPECsfs97.v3 ops per second with an overall response time of 4.93 ms	4000	88	45.45	3072	1560000 (39)	1760	1936 (92%)	122672+0	0	122672+0	0	I/O working set was cacheable; probably not affected
3991 SPECsfs97.v3 ops per second with an overall response time of 3.50 ms	4008	36	111.33	1056	1563120 (39)	1440	972 (75%)	83160+8460	8460	83160+8460	176796	probably affected
4043 SPECsfs97.v3 ops per second with an overall response time of 6.86 ms	4020	126	31.90	1216	1567800 (39)	2016	1512 (100%)	95760+35910	66780	95760+35910	104076	probably affected
4092 SPECsfs97.v3 ops per second with an overall response time of 2.70 ms	4098	30	136.60	1056	1598220 (40)	1440	870 (60%)	83160+2310	2310	83160+2310	193680	probably affected
4174 SPECsfs97.v3 ops per second with an overall response time of 6.94 ms	4180	128	32.66	3584	1630200 (40)	2048	1536 (100%)	135808+0	0	135808+0	0	I/O working set was cacheable; probably not affected
4219 SPECsfs97.v3 ops per second with an overall response time of 3.74 ms	4200	180	23.33	7168	1638000 (41)	2160	2160 (100%)	146340+0	0	146340+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
4380 SPECsfs97.v3 ops per second with an overall response time of 3.65 ms	4380	48	91.25	1056	1708200 (42)	1728	1200 (69%)	83136+19680	30192	83136+19680	242832	probably affected
4606 SPECsfs97.v3 ops per second with an overall response time of 3.80 ms	4560	216	21.11	7168	1778400 (44)	2592	2592 (100%)	157464+0	0	157464+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
4616 SPECsfs97.v3 ops per second with an overall response time of 2.66 ms	4650	30	155.00	1056	1813500 (45)	1680	1020 (57%)	83160+6120	6120	83160+6120	303540	probably affected
4709 SPECsfs97.v3 ops per second with an overall response time of 6.56 ms	4800	112	42.86	3072	1872000 (46)	2240	2464 (92%)	146160+0	0	146160+0	0	I/O working set was cacheable; probably not affected
5011 SPECsfs97.v3 ops per second with an overall response time of 3.71 ms	5000	64	78.12	512	1950000 (48)	2048	1728 (75%)	40320+84544	566976	40320+84544	826368	probably affected
5191 SPECsfs97.v3 ops per second with an overall response time of 3.58 ms	5200	64	81.25	512	2028000 (50)	2048	1728 (75%)	40320+88896	614464	40320+88896	883712	probably affected
5382 SPECsfs97.v3 ops per second with an overall response time of 2.48 ms	5400	54	100.00	4960	2106000 (52)	1944	1350 (69%)	128304+0	0	128304+0	0	I/O working set was cacheable; probably not affected
5412 SPECsfs97.v3 ops per second with an overall response time of 6.30 ms	5632	112	50.29	3072	2196480 (55)	2240	2464 (92%)	172928+0	0	172928+0	0	I/O working set was cacheable; probably not affected
5484 SPECsfs97.v3 ops per second with an overall response time of 3.69 ms	5500	64	85.94	512	2145000 (53)	2048	1728 (75%)	40320+95680	677376	40320+95680	957120	probably affected
5597 SPECsfs97.v3 ops per second with an overall response time of 3.54 ms	5600	64	87.50	512	2184000 (54)	2048	1728 (75%)	40320+98816	705280	40320+98816	990080	probably affected
5602 SPECsfs97.v3 ops per second with an overall response time of 5.63 ms	5600	84	66.67	3072	2184000 (54)	2352	1764 (88%)	148092+0	0	148092+0	0	I/O working set was cacheable; probably not affected
5894 SPECsfs97.v3 ops per second with an overall response time of 2.96 ms	5900	128	46.09	1024	2301000 (57)	2560	2816 (92%)	80640+100224	414976	80640+100224	594688	probably affected
5903 SPECsfs97.v3 ops per second with an overall response time of 5.41 ms	5940	120	49.50	3072	2316600 (58)	2400	2640 (92%)	181800+0	0	181800+0	0	I/O working set was cacheable; probably not affected
6011 SPECsfs97.v3 ops per second with an overall response time of 4.70 ms	6050	132	45.83	4096	2359500 (59)	2640	2904 (92%)	186120+0	0	186120+0	0	I/O working set was cacheable; probably not affected
6018 SPECsfs97.v3 ops per second with an overall response time of 4.97 ms	6264	128	48.94	6400	2442960 (61)	2560	2816 (92%)	193152+0	0	193152+0	0	I/O working set was cacheable; probably not affected
6267 SPECsfs97.v3 ops per second with an overall response time of 1.79 ms	6600	20	330.00	1088	2574000 (64)	2160	820 (38%)	85700+7100	8780	85700+7100	720260	probably affected
6277 SPECsfs97.v3 ops per second with an overall response time of 6.21 ms	6300	84	75.00	3072	2457000 (61)	2352	1764 (88%)	168924+0	0	168924+0	0	I/O working set was cacheable; probably not affected
6573 SPECsfs97.v3 ops per second with an overall response time of 2.57 ms	6534	54	121.00	4960	2548260 (63)	2376	1620 (62%)	141912+0	0	141912+0	0	I/O working set was cacheable; probably not affected
6589 SPECsfs97.v3 ops per second with an overall response time of 3.58 ms	6600	96	68.75	5120	2574000 (64)	2688	2016 (88%)	176544+0	0	176544+0	0	I/O working set was cacheable; probably not affected
6594 SPECsfs97.v3 ops per second with an overall response time of 6.41 ms	6600	168	39.29	5120	2574000 (64)	3360	3696 (92%)	201432+0	0	201432+0	0	I/O working set was cacheable; probably not affected
6616 SPECsfs97.v3 ops per second with an overall response time of 3.00 ms	6600	128	51.56	1024	2574000 (64)	3072	2688 (88%)	80640+106496	485376	80640+106496	762880	probably affected
6745 SPECsfs97.v3 ops per second with an overall response time of 4.83 ms	6765	132	51.25	4096	2638350 (66)	2640	2904 (92%)	206844+0	0	206844+0	0	I/O working set was cacheable; probably not affected
6852 SPECsfs97.v3 ops per second with an overall response time of 5.54 ms	7040	128	55.00	6400	2745600 (68)	3072	2688 (88%)	202496+0	0	202496+0	0	I/O working set was cacheable; probably not affected
7248 SPECsfs97.v3 ops per second with an overall response time of 1.96 ms	7380	24	307.50	1088	2878200 (72)	2400	960 (42%)	85680+21168	44760	85680+21168	906264	probably affected
7296 SPECsfs97.v3 ops per second with an overall response time of 6.18 ms	7320	168	43.57	5120	2854800 (71)	3360	3696 (92%)	221928+0	0	221928+0	0	I/O working set was cacheable; probably not affected
7301 SPECsfs97.v3 ops per second with an overall response time of 3.70 ms	7500	288	26.04	14336	2925000 (73)	4608	3456 (100%)	248832+0	0	248832+0	0	I/O working set was cacheable; probably not affected
7383 SPECsfs97.v3 ops per second with an overall response time of 1.54 ms	7340	20	367.00	1152	2862600 (71)	2400	860 (36%)	90740+6860	7420	90740+6860	849020	probably affected
7428 SPECsfs97.v3 ops per second with an overall response time of 3.09 ms	7548	72	104.83	2112	2943720 (73)	2880	1944 (75%)	166320+6912	6912	166320+6912	273960	probably affected
7745 SPECsfs97.v3 ops per second with an overall response time of 1.70 ms	7735	28	276.25	2112	3016650 (75)	2576	1176 (44%)	119560+0	0	119560+0	309036	probably affected
7783 SPECsfs97.v3 ops per second with an overall response time of 1.55 ms	8100	24	337.50	3200	3159000 (79)	2688	1008 (39%)	112128+0	0	112128+0	36528	probably affected
7900 SPECsfs97.v3 ops per second with an overall response time of 1.66 ms	7850	20	392.50	1152	3061500 (76)	2560	960 (36%)	90740+11040	14840	90740+11040	972780	probably affected
8011 SPECsfs97.v3 ops per second with an overall response time of 6.26 ms	8000	112	71.43	5120	3120000 (78)	3136	2352 (88%)	214032+0	0	214032+0	0	I/O working set was cacheable; probably not affected
8028 SPECsfs97.v3 ops per second with an overall response time of 2.54 ms	8000	64	125.00	512	3120000 (78)	2816	1920 (62%)	40320+132096	1225088	40320+132096	1675072	probably affected
8030 SPECsfs97.v3 ops per second with an overall response time of 3.16 ms	8196	72	113.83	2112	3196440 (80)	2880	1944 (75%)	166320+21240	21240	166320+21240	387504	probably affected
8067 SPECsfs97.v3 ops per second with an overall response time of 3.76 ms	8160	432	18.89	14336	3182400 (79)	5184	5184 (100%)	282960+0	0	282960+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
8131 SPECsfs97.v3 ops per second with an overall response time of 6.91 ms	8160	168	48.57	5120	3182400 (79)	3360	3696 (92%)	248976+0	0	248976+0	0	I/O working set was cacheable; probably not affected
8503 SPECsfs97.v3 ops per second with an overall response time of 4.65 ms	8730	198	44.09	6144	3404700 (85)	3960	4356 (92%)	267300+0	0	267300+0	0	I/O working set was cacheable; probably not affected
8637 SPECsfs97.v3 ops per second with an overall response time of 6.13 ms	8640	112	77.14	5120	3369600 (84)	3584	3024 (75%)	215936+0	0	215936+0	0	I/O working set was cacheable; probably not affected
8787 SPECsfs97.v3 ops per second with an overall response time of 2.24 ms	8750	64	136.72	512	3412500 (85)	3072	1856 (60%)	40320+144128	1396800	40320+144128	1883136	probably affected
8798 SPECsfs97.v3 ops per second with an overall response time of 5.72 ms	8892	117	76.00	8192	3467880 (86)	3276	2457 (88%)	237159+0	0	237159+0	0	I/O working set was cacheable; probably not affected
8819 SPECsfs97.v3 ops per second with an overall response time of 1.84 ms	9000	32	281.25	2112	3510000 (87)	2944	1344 (44%)	138752+0	0	138752+0	539904	probably affected
8820 SPECsfs97.v3 ops per second with an overall response time of 1.98 ms	8820	36	245.00	3200	3439800 (86)	3024	1332 (44%)	140760+0	0	140760+0	89604	probably affected
9073 SPECsfs97.v3 ops per second with an overall response time of 6.24 ms	9240	168	55.00	5120	3603600 (90)	4032	3528 (88%)	265776+0	0	265776+0	0	I/O working set was cacheable; probably not affected
9212 SPECsfs97.v3 ops per second with an overall response time of 4.20 ms	9270	198	46.82	6144	3615300 (90)	3960	4356 (92%)	283932+0	0	283932+0	0	I/O working set was cacheable; probably not affected
9753 SPECsfs97.v3 ops per second with an overall response time of 5.99 ms	9760	144	67.78	7168	3806400 (95)	4032	3024 (88%)	260640+0	0	260640+0	0	I/O working set was cacheable; probably not affected
9935 SPECsfs97.v3 ops per second with an overall response time of 4.84 ms	9900	180	55.00	7168	3861000 (96)	4320	3780 (88%)	284760+0	0	284760+0	0	I/O working set was cacheable; probably not affected
10011 SPECsfs97.v3 ops per second with an overall response time of 1.24 ms	10080	24	420.00	3200	3931200 (98)	3264	1104 (35%)	126816+0	0	126816+0	232176	probably affected
10059 SPECsfs97.v3 ops per second with an overall response time of 3.53 ms	10000	128	78.12	1024	3900000 (97)	4096	3456 (75%)	80640+169088	1133952	80640+169088	1652736	probably affected
10335 SPECsfs97.v3 ops per second with an overall response time of 3.29 ms	10400	128	81.25	1024	4056000 (101)	4096	3456 (75%)	80640+177792	1228928	80640+177792	1767424	probably affected
10390 SPECsfs97.v3 ops per second with an overall response time of 5.28 ms	10664	192	55.54	8576	4158960 (104)	4608	4032 (88%)	307392+0	0	307392+0	0	I/O working set was cacheable; probably not affected
10422 SPECsfs97.v3 ops per second with an overall response time of 3.10 ms	10500	180	58.33	2036	4095000 (102)	4320	3780 (88%)	160200+138600	470520	160200+138600	901260	probably affected
10497 SPECsfs97.v3 ops per second with an overall response time of 3.34 ms	10500	128	82.03	1024	4095000 (102)	4096	3456 (75%)	80640+178048	1245312	80640+178048	1792384	probably affected
10517 SPECsfs97.v3 ops per second with an overall response time of 3.81 ms	10560	480	22.00	14336	4118400 (103)	5760	5760 (100%)	365280+0	0	365280+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
10592 SPECsfs97.v3 ops per second with an overall response time of 5.66 ms	10560	192	55.00	6144	4118400 (103)	4608	4032 (88%)	303744+0	0	303744+0	0	I/O working set was cacheable; probably not affected
10612 SPECsfs97.v3 ops per second with an overall response time of 4.31 ms	10800	198	54.55	8192	4212000 (105)	4752	4158 (88%)	309870+0	0	309870+0	0	I/O working set was cacheable; probably not affected
10678 SPECsfs97.v3 ops per second with an overall response time of 6.20 ms	10800	168	64.29	9216	4212000 (105)	4704	3528 (88%)	286104+0	0	286104+0	0	I/O working set was cacheable; probably not affected
10975 SPECsfs97.v3 ops per second with an overall response time of 3.29 ms	11000	128	85.94	1024	4290000 (107)	4096	3456 (75%)	80640+191360	1354752	80640+191360	1914240	probably affected
11275 SPECsfs97.v3 ops per second with an overall response time of 3.35 ms	11168	768	14.54	14336	4355520 (109)	9216	9216 (100%)	388608+0	0	388608+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
11322 SPECsfs97.v3 ops per second with an overall response time of 2.50 ms	12216	48	254.50	10240	4764240 (119)	4032	1776 (44%)	195744+0	0	195744+0	0	I/O working set was cacheable; probably not affected
11451 SPECsfs97.v3 ops per second with an overall response time of 3.15 ms	12100	90	134.44	2036	4719000 (118)	4320	2610 (60%)	160290+93690	286920	160290+93690	1285200	probably affected
11452 SPECsfs97.v3 ops per second with an overall response time of 3.89 ms	11376	180	63.20	8192	4436640 (111)	4320	3780 (88%)	323640+0	0	323640+0	0	I/O working set was cacheable; probably not affected
11454 SPECsfs97.v3 ops per second with an overall response time of 4.43 ms	11550	264	43.75	8192	4504500 (112)	5280	5808 (92%)	355344+0	0	355344+0	0	I/O working set was cacheable; probably not affected
11489 SPECsfs97.v3 ops per second with an overall response time of 4.32 ms	11448	180	63.60	7168	4464720 (111)	4320	3780 (88%)	327780+0	0	327780+0	0	I/O working set was cacheable; probably not affected
11626 SPECsfs97.v3 ops per second with an overall response time of 1.44 ms	11600	32	362.50	3200	4524000 (113)	3840	1376 (36%)	153792+0	0	153792+0	444480	probably affected
11636 SPECsfs97.v3 ops per second with an overall response time of 3.10 ms	11696	256	45.69	2048	4561440 (114)	5120	5632 (92%)	161280+199680	810496	161280+199680	1163520	probably affected
11716 SPECsfs97.v3 ops per second with an overall response time of 5.88 ms	11990	240	49.96	8576	4676100 (117)	4800	5280 (92%)	366240+0	0	366240+0	0	I/O working set was cacheable; probably not affected
12121 SPECsfs97.v3 ops per second with an overall response time of 4.22 ms	12100	450	26.89	3648	4719000 (118)	7200	5400 (100%)	287100+108450	204750	287100+108450	324450	probably affected
12121 SPECsfs97.v3 ops per second with an overall response time of 5.54 ms	12288	224	54.86	12288	4792320 (120)	5376	4704 (88%)	353024+0	0	353024+0	0	I/O working set was cacheable; probably not affected
12231 SPECsfs97.v3 ops per second with an overall response time of 2.72 ms	12984	48	270.50	10240	5063760 (126)	4416	2016 (44%)	200016+0	0	200016+0	0	I/O working set was cacheable; probably not affected
12664 SPECsfs97.v3 ops per second with an overall response time of 3.54 ms	12600	450	28.00	3648	4914000 (123)	7200	5400 (100%)	287100+130500	273150	287100+130500	417600	probably affected
12809 SPECsfs97.v3 ops per second with an overall response time of 4.44 ms	13200	264	50.00	12288	5148000 (128)	5280	5808 (92%)	403392+0	0	403392+0	0	I/O working set was cacheable; probably not affected
12836 SPECsfs97.v3 ops per second with an overall response time of 5.39 ms	12800	224	57.14	12288	4992000 (125)	5376	4704 (88%)	368928+0	0	368928+0	0	I/O working set was cacheable; probably not affected
12983 SPECsfs97.v3 ops per second with an overall response time of 4.36 ms	12960	180	72.00	8192	5054400 (126)	5040	3780 (88%)	348660+0	0	348660+0	0	I/O working set was cacheable; probably not affected
13021 SPECsfs97.v3 ops per second with an overall response time of 3.33 ms	13000	256	50.78	2048	5070000 (126)	5120	5632 (92%)	161280+239104	1084160	161280+239104	1470720	probably affected
13056 SPECsfs97.v3 ops per second with an overall response time of 1.46 ms	13000	32	406.25	2304	5070000 (126)	4224	1440 (34%)	165024+0	0	165024+0	1292800	probably affected
13795 SPECsfs97.v3 ops per second with an overall response time of 4.51 ms	13860	264	52.50	12288	5405400 (135)	6336	5544 (88%)	396528+0	0	396528+0	0	I/O working set was cacheable; probably not affected
14065 SPECsfs97.v3 ops per second with an overall response time of 5.51 ms	14000	224	62.50	12288	5460000 (136)	5376	4704 (88%)	400960+0	0	400960+0	0	I/O working set was cacheable; probably not affected
14157 SPECsfs97.v3 ops per second with an overall response time of 3.14 ms	14196	216	65.72	16384	5536440 (138)	6048	4536 (88%)	376488+0	0	376488+0	0	I/O working set was cacheable; probably not affected
14186 SPECsfs97.v3 ops per second with an overall response time of 1.55 ms	14304	32	447.00	2304	5578560 (139)	4608	1504 (33%)	174816+0	0	174816+0	1603968	probably affected
14894 SPECsfs97.v3 ops per second with an overall response time of 5.65 ms	15120	224	67.50	12288	5896800 (147)	6272	4704 (88%)	405440+0	0	405440+0	0	I/O working set was cacheable; probably not affected
14934 SPECsfs97.v3 ops per second with an overall response time of 4.41 ms	14850	264	56.25	12288	5791500 (145)	6336	5544 (88%)	425304+0	0	425304+0	0	I/O working set was cacheable; probably not affected
15284 SPECsfs97.v3 ops per second with an overall response time of 4.92 ms	15360	240	64.00	12288	5990400 (150)	5760	5040 (88%)	444960+0	0	444960+0	0	I/O working set was cacheable; probably not affected
15723 SPECsfs97.v3 ops per second with an overall response time of 2.73 ms	15748	128	123.03	1024	6141720 (153)	5632	3840 (62%)	80640+260480	2408064	80640+260480	3270272	probably affected
16020 SPECsfs97.v3 ops per second with an overall response time of 4.60 ms	15950	264	60.42	12288	6220500 (155)	6336	5544 (88%)	459360+0	0	459360+0	0	I/O working set was cacheable; probably not affected
16049 SPECsfs97.v3 ops per second with an overall response time of 3.96 ms	16320	864	18.89	21504	6364800 (159)	10368	10368 (100%)	565920+0	0	565920+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
16547 SPECsfs97.v3 ops per second with an overall response time of 3.12 ms	16500	168	98.21	8192	6435000 (161)	6048	4200 (69%)	390264+0	0	390264+0	0	I/O working set was cacheable; probably not affected
16557 SPECsfs97.v3 ops per second with an overall response time of 2.99 ms	16500	168	98.21	8192	6435000 (161)	6048	4200 (69%)	390264+0	0	390264+0	0	I/O working set was cacheable; probably not affected
16913 SPECsfs97.v3 ops per second with an overall response time of 5.10 ms	17904	240	74.60	12288	6982560 (174)	6720	5040 (88%)	479760+0	0	479760+0	0	I/O working set was cacheable; probably not affected
16997 SPECsfs97.v3 ops per second with an overall response time of 3.39 ms	16992	360	47.20	3072	6626880 (165)	7200	7920 (92%)	241920+283680	1118880	241920+283680	1621080	probably affected
17087 SPECsfs97.v3 ops per second with an overall response time of 4.83 ms	17655	396	44.58	18432	6885450 (172)	7920	8712 (92%)	542124+0	0	542124+0	0	I/O working set was cacheable; probably not affected
17286 SPECsfs97.v3 ops per second with an overall response time of 1.33 ms	17710	44	402.50	6400	6906900 (172)	5808	1980 (34%)	223696+0	0	223696+0	183216	probably affected
17542 SPECsfs97.v3 ops per second with an overall response time of 4.36 ms	17568	240	73.20	12288	6851520 (171)	6720	5040 (88%)	469200+0	0	469200+0	0	I/O working set was cacheable; probably not affected
17766 SPECsfs97.v3 ops per second with an overall response time of 3.75 ms	17760	1728	10.28	21504	6926400 (173)	13824	19008 (92%)	656640+0	0	656640+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
18535 SPECsfs97.v3 ops per second with an overall response time of 4.25 ms	18876	396	47.67	18432	7361640 (184)	7920	8712 (92%)	578160+0	0	578160+0	0	I/O working set was cacheable; probably not affected
18792 SPECsfs97.v3 ops per second with an overall response time of 3.92 ms	19200	352	54.55	24576	7488000 (187)	8448	7392 (88%)	550880+0	0	550880+0	0	I/O working set was cacheable; probably not affected
18982 SPECsfs97.v3 ops per second with an overall response time of 3.53 ms	18996	360	52.77	3072	7408440 (185)	8640	7560 (88%)	241920+298800	1285920	241920+298800	2074320	probably affected
19520 SPECsfs97.v3 ops per second with an overall response time of 1.46 ms	19600	56	350.00	6400	7644000 (191)	6496	2576 (38%)	267400+0	0	267400+0	383096	probably affected
19886 SPECsfs97.v3 ops per second with an overall response time of 4.49 ms	20160	240	84.00	15360	7862400 (196)	7680	6480 (75%)	499440+0	0	499440+0	0	I/O working set was cacheable; probably not affected
21476 SPECsfs97.v3 ops per second with an overall response time of 4.03 ms	21440	352	60.91	24576	8361600 (209)	8448	7392 (88%)	612480+0	0	612480+0	0	I/O working set was cacheable; probably not affected
22701 SPECsfs97.v3 ops per second with an overall response time of 3.77 ms	22736	512	44.41	4096	8867040 (222)	10240	11264 (92%)	322560+378368	1476096	322560+378368	2212352	probably affected
23963 SPECsfs97.v3 ops per second with an overall response time of 3.70 ms	24000	512	46.88	4096	9360000 (234)	10240	11264 (92%)	322560+411648	1731072	322560+411648	2451456	probably affected
25565 SPECsfs97.v3 ops per second with an overall response time of 3.16 ms	25500	360	70.83	44032	9945000 (249)	10080	7560 (88%)	676080+0	0	676080+0	0	I/O working set was cacheable; probably not affected
29083 SPECsfs97.v3 ops per second with an overall response time of 4.25 ms	29000	240	120.83	16384	11310000 (283)	10560	7200 (62%)	629760+0	0	629760+0	0	I/O working set was cacheable; probably not affected
29997 SPECsfs97.v3 ops per second with an overall response time of 3.91 ms	30000	2400	12.50	44032	11700000 (292)	19200	26400 (92%)	1096800+0	0	1096800+0	0	<26 requested ops/sec/process and I/O working set was cacheable; probably not affected
32049 SPECsfs97.v3 ops per second with an overall response time of 2.62 ms	32000	256	125.00	2048	12480000 (312)	11264	7680 (62%)	161280+528384	4900352	161280+528384	6700288	probably affected
34913 SPECsfs97.v3 ops per second with an overall response time of 3.89 ms	35000	784	44.64	7168	13650000 (341)	15680	17248 (92%)	564480+516656	1697360	564480+516656	2743216	probably affected
37909 SPECsfs97.v3 ops per second with an overall response time of 4.04 ms	37996	784	48.46	7168	14818440 (371)	15680	17248 (92%)	564480+597408	2228912	564480+597408	3390800	probably affected
61809 SPECsfs97.v3 ops per second with an overall response time of 2.93 ms	62000	512	121.09	4096	24180000 (605)	22528	14336 (58%)	322560+1022976	9334784	322560+1022976	12857856	probably affected
104067 SPECsfs97.v3 ops per second with an overall response time of 2.87 ms	105000	784	133.93	7168	40950000 (1025)	37632	22736 (60%)	564480+1629152	14666288	564480+1629152	21246400	probably affected
120206 SPECsfs97.v3 ops per second with an overall response time of 1.87 ms	120000	768	156.25	51200	46800000 (1171)	43008	26112 (57%)	2314752+0	0	2314752+0	148224	probably affected