Fujitsu SPARC M10-4S CODE DEFECT (CD) RESULT
Tested by: Fujitsu Test Sponsor: Fujitsu Test location: Numazu, Japan Test date: Apr 25, 2014
SPEC license #: 19 Hardware Availability: Apr-2014 Software Availability: Apr-2014 Publication: May 15, 2014
Validation Errors Summary
SPECjbb2013 results have been marked as "Code Defect" due to a bug that impacts the comparability of results. The bug can significantly reduce the amount of work done during the measurement period, resulting in an inflated SPECjbb2013 metric.
Benchmark Results Summary
 
Overall Throughput RT curve
Overall SUT (System Under Test) Description
VendorFujitsu
Vendor URLhttp://www.fujitsu.com/
System SourceSingle Supplier
System DesignationServer Rack
Total Systems1
All SUT Systems IdenticalYES
Total Nodes1
All Nodes IdenticalYES
Nodes Per System1
Total Chips16
Total Cores256
Total Threads512
Total Memory Amount (GB)8192
Total OS Images1
SW EnvironmentNon-virtual
 
Hardware hw_1
NameSPARC M10-4S
VendorFujitsu
Vendor URLhttp://www.fujitsu.com/
AvailableApr-2014
ModelSPARC M10-4S
Form Factor4RU rackmount server
CPU NameSPARC64 X+
CPU Characteristics16 Core, 3.7GHz, 24MB L2 Cache
Number of Systems1
Nodes Per System1
Chips Per System16
Cores Per System256
Cores Per Chip16
Threads Per System512
Threads Per Core2
VersionNone
CPU Frequency (MHz)3700
Primary Cache64 KB I + 64 KB D on chip per core
Secondary Cache24 MB I+D on chip per chip
Tertiary CacheNone
Other CacheNone
Disk1 x 600 GB SAS, 10025 RPM SAS disk drive
File SystemZFS
Memory Amount (GB)8192
# and size of DIMM(s)256 x 32 GB
Memory Details32GB 4Rx4 PC3-10600R-9 ECC; All slots populated
# and type of Network Interface Cards (NICs)4 x 10/100/1000 NICs (on board)
Power Supply Quantity and Rating (W)8 x 2500
Other HardwareNone
Cabinet/Housing/EnclosureNone
Shared DescriptionNone
Shared CommentNone
NotesNone
Other Hardware network_1
NameNone
VendorNone
Vendor URLNone
VersionNone
AvailableNone
BitnessNone
NotesNone
Operating System os_1
NameOracle Solaris 11.1 SRU 15.4
VendorOracle Corporation
Vendor URLhttp://www.oracle.com/
VersionOracle Solaris 11.1 SRU 15.4
AvailableFeb-2014
Bitness64
NotesNone
Java Virtual Machine jvm_1
NameOracle Java SE 8u5
VendorOracle Corporation
Vendor URLhttp://www.oracle.com/
VersionJava HotSpot(TM) 64-Bit Server VM, version 1.8.0_05
AvailableApr-2014
Bitness64
NotesNone
Other Software other_1
NameNone
VendorNone
Vendor URLNone
VersionNone
AvailableNone
BitnessNone
NotesNone
Hardware
OS Images os_Image_1(1)
Hardware Description hw_1
Number of Systems 1
SW Environment Non-virtual
Tuning None
Notes None
OS Image os_Image_1
JVM Instances jvm_Ctr_1(1), jvm_Backend_1(16), jvm_TxInjector_1(16)
OS Image Description os_1
Tuning /etc/system parameters
autoup = 1555200
Causes pages older than the listed number of seconds to be written by fsflush.
tune_t_fsflushr = 259200
Controls how many seconds elapse between runs of the page flush daemon, fsflush.
Notes All java processes were run using the FX scheduling class. The "Logical Domains Manager" service was turned off using svcadm disable ldmd.
JVM Instance jvm_Ctr_1
Parts of Benchmark Controller
JVM Instance Description jvm_1
Command Line -Xmx2000m -Xms2000m -Xmn1200m -XX:LargePageSizeInBytes=4m -XX:MaxInlineSize=200 -XX:InlineSmallCode=2000 -verbose:gc -XX:ParallelGCThreads=4
Tuning Conroller was run outside of processor sets.
Notes The object-caching memory allocation library libumem was used by setting the environment variable LD_PRELOAD_64=/lib/sparcv9/libumem.so.
JVM Instance jvm_Backend_1
Parts of Benchmark Backend
JVM Instance Description jvm_1
Command Line -verbose:gc -XX:+PrintGCDetails -XX:+AlwaysPreTouch -XX:LargePageSizeInBytes=256m -XX:AllocatePrefetchLines=8 -XX:InlineSmallCode=2000 -XX:MaxInlineSize=300 -Xms320g -Xmx320g -Xmn300g -XX:+UseParallelOldGC -XX:-UseAdaptiveSizePolicy -XX:ParallelGCThreads=32 -XX:SurvivorRatio=100 -XX:MaxTenuringThreshold=1
Tuning Each group (backend and transaction injector) is run in its own processor set consisting of all the virtual processors (threads) belonging to the same physical CPU.
  • $GROUPID < $GROUP_COUNT:
    • psrset -c <all virtual processors of physical CPU#$GROUPID-1>
    • psrset -e $GROUPID $JAVA $JAVA_OPTS_BE -jar specjbb2013.jar -m BACKEND -G=$GROUPID -J=$JVMID
  • $GROUPID = $GROUP_COUNT:
    • $JAVA $JAVA_OPTS_BE -jar specjbb2013.jar -m BACKEND -G=$GROUPID -J=$JVMID
Notes The object-caching memory allocation library libumem was used by setting the environment variable LD_PRELOAD_64=/lib/sparcv9/libumem.so. All java processes were run using the FX scheduling class.
JVM Instance jvm_TxInjector_1
Parts of Benchmark TxInjector
JVM Instance Description jvm_1
Command Line -Xmx2400m -Xms2400m -Xmn1600m -verbose:gc -XX:+PrintGCDetails -XX:ParallelGCThreads=2 -XX:+UseConcMarkSweepGC -XX:ConcGCThreads=1 -XX:SurvivorRatio=8 -XX:TargetSurvivorRatio=90 -XX:MaxTenuringThreshold=8
Tuning Each group (backend and transaction injector) is run in its own processor set consisting of all the virtual processors (threads) belonging to the same physical CPU.
  • $GROUPID < $GROUP_COUNT:
    • psrset -c <all virtual processors of physical CPU#$GROUPID-1>
    • psrset -e $GROUPID $JAVA $JAVA_OPTS_BE -jar specjbb2013.jar -m BACKEND -G=$GROUPID -J=$JVMID
  • $GROUPID = $GROUP_COUNT:
    • $JAVA $JAVA_OPTS_BE -jar specjbb2013.jar -m BACKEND -G=$GROUPID -J=$JVMID
Notes The object-caching memory allocation library libumem was used by setting the environment variable LD_PRELOAD_64=/lib/sparcv9/libumem.so. All java processes were run using the FX scheduling class.
max-jOPS = jOPS passed before the First Failure
Pass/Fail Pass Pass Fail Fail Fail
jOPS 420774 425348 429921 434495 439068
critical-jOPS = Geomean ( jOPS @ 10000; 50000; 100000; 200000; 500000; SLAs )
Response time percentile is 99-th
SLA (us) 10000 50000 100000 200000 500000 Geomean
jOPS 107480 198495 213436 240115 281278 198418
  Percentile
  10-th 50-th 90-th 95-th 99-th 100-th
500us - / 4574 - / 4574 - / 4574 - / 4574 - / 4574 - / 4574
1000us 13721 / 18295 4574 / 9147 - / 4574 - / 4574 - / 4574 - / 4574
5000us 233255 / 237829 160077 / 155503 73178 / 77752 68604 / 73178 54884 / 59457 - / 4574
10000us 246976 / 251550 224108 / 228681 178372 / 182945 160077 / 155503 114341 / 100620 - / 4574
20000us 265271 / 269844 233255 / 237829 224108 / 228681 205813 / 210387 178372 / 164651 - / 4574
50000us 278991 / 283565 242402 / 246976 228681 / 233255 224108 / 228681 205813 / 187519 27442 / 4574
100000us 311007 / 315580 260697 / 265271 237829 / 242402 233255 / 237829 219534 / 210387 68604 / 4574
200000us 333875 / 338449 288139 / 292712 251550 / 256123 246976 / 251550 237829 / 242402 150930 / 32015
500000us 375038 / 379611 352170 / 356743 311007 / 315580 297286 / 301860 278991 / 283565 219534 / 59457
1000000us 407053 / 411627 384185 / 388759 370464 / 375038 361317 / 365890 347596 / 343022 242402 / 228681
Probes jOPS / Total jOPS
Request Mix Accuracy
Note
(Actual % in the Mix - Expected % in the Mix) must be within:
'Main Tx' limit of +/-5.0% for the requests whose expected % in the mix is >= 10.0%
'Minor Tx' limit of +/-1.0% for the requests whose expected % in the mix is < 10.0%
There were no non-critical failures in Response Time curve building
Delay between status pings
IR/PR Accuracy
This section lists properties only set by user
Property Name Default Controller
specjbb.group.count 1 16
specjbb.txi.pergroup.count 1 1
View table in csv format
 
Level: COMPLIANCE
Check Agent Result
Check properties on compliance All PASSED
 
Level: CORRECTNESS
Check Agent Result
Compare SM and HQ Inventory All PASSED
High-bound (max attempted) is 457363 IR
High-bound (settled) is 381403 IR