Oracle hardwires encryption and SQL hastening algorithms into Sparc M7 silicon Oracle execs used the final keynote of this week's OpenWorld to praise their Sparc M7 processor's ability to accelerate encryption and some SQL queries in hardware. On Wednesday, John Fowler, veep of systems at Oracle, said the M7 microprocessor and its builtin coprocessors that speed up crypto algorithms and database requests …
The application for decompression goes beyond straight Oracle RDBMS.
- Live migration of VM's (LDom's & Zones) between Hypervisors
- Speed up Open Source databases sitting on top of compressed ZFS filesystem
- Larger bandwidth of normal data communicating from a back-end system to a Cloud provider
- Reduced cost for syncronization between storage at diverse locations using hardware decompression
When compression/decompression is tied with hardware encryption/decryption - secure & cheap is hard to beat!
Solaris planned the compression/decompression prior the encryption/decryption engines in components like ZFS... so this has been getting done in software, for a long time. Nice to get a boost in CPU power AND get a higher percentage of that CPU back for application work!
SAP S&D Tier 2 benchmark delivers 481 Users per core on the SPARC T7 compared to 996 with IBM's POWER8. Memory bandwidth of 160 GB/s per socket with 4 memory controllers servicing 32 cores versus 230 GB/s per socket with 2 memory controllers servicing 12 x POWER8 cores. Speaks for itself.
Hardly impressive on SAP vs Power8???
Well you need three (3) IBM Power8 CPUs to match one single Sparc M7 in SAP benchmarks. If you consider power8 CPU to be fast, well the M7 CPU is three times faster. Isn't that impressive? Which CPU is the fastest in the world today? Sparc M7, beating everyone. It turns out that Intel Xeon is faster than the latest and newest power8 in some benchmarks, so power8 is worst in class, rendering it obsolete.
If memory bandwidth would be a problem, the M7 could not be 2-3 faster than power8 in benchmarks. For instance you need more than two power8 CPUs to match the M7 in spec2006 CPU benchmarks. How is that possible if bandwidth is a problem? IBM has lost big time, worst in class. IBM better exit the CPU market with these inferior and slow products
Here are some Sparc M7 world records. It is typically 3x faster than the fastest x86 v3 CPU, and IBM power8. Going all the way to being >10x faster for database workloads. Btw, it is almost twice as fast as IBM power8 on STREAM memory bandwidth too.
https://blogs.oracle.com/BestPerf/
@PowerMan
"...[Sparc M7] Memory bandwidth of 160 GB/s per socket with 4 memory controllers servicing 32 cores versus 230 GB/s per socket with 2 memory controllers servicing 12 x POWER8 cores. Speaks for itself...."
Well, mr PowerMan, here are some information about IBMs superior memory bandwidth POWER8 claims. So, maybe you should not believe the IBM marketing FUD. It seems that Oracle SPARC M7 160GB/sec is faster than IBM Power8 230GB/sec in practice:
https://blogs.oracle.com/BestPerf/entry/20151025_stream_sparcm7
"...IBM says the sustained or delivered bandwidth of the IBM POWER8 12-core chip is 230 GB/s. This number is a peak bandwidth calculation: 230.4 GB/sec = 9.6 GHz * 3 (r+w) * 8 byte. A similar calculation is used by IBM for the POWER8 dual-chip-module (two 6-core chips) to show a sustained or delivered bandwidth of 192 GB/sec (192.0 GB/sec = 8.0 GHz * 3 (r+w) * 8 byte). Peaks are the theoretical limits used for marketing hype, but true measured delivered bandwidth is the only useful comparison to help one understand delivered performance of real applications.
...
The SPARC T7-4 server delivered over... 2.3 times the triad bisection bandwidth of a four-chip IBM Power System S824 server...."
Fast relates to Speed which relates to Response Time.
Throughput related to Volume
A Ferrari is Fast, takes 2 people from point A to B point B really fast
A Bus has Throughput, it takes 40 people from A to B at a time, but at a reduced speed
A 32 core processor has a high Throughput. It can do a bunch of stuff at the same time but it doesn't mean it is fast.
According to SAP SD 2-Tier benchmark of the T7-2, response times for dialog/updates where 0.022 sec / 0.047 sec. The values for the POWER8 where 0.011 sec / 0.019 sec.
T7 is a large processor, that is a fact. With 32 cores any comparison "per processor" will lead to a clear advantage. Sadly, licensing isn't usually measured by Silicon chip but by cores.
Having the fastest per core performance will lead to savings.
Having the largest processor will allow you to brag with your friends.
Yes, true, having the fastest per core performance will lead to savings - especially Database license savings when their priced per core. Unfortunately the SAP benchmark is an application benchmark so not a good indicator of Database performance.
Oracle recently ran the open source Database testing tool, HammerDB which simulates a TPC-C OLTP workload, and has published results comparing to Intel Xeon E5-v3 and even IBM's Power8 based S824. A single chip SPARC T7-1 (yes, with 32-cores), outperforms the Power S824 with 24-cores (with 4 x Power8 6-core chips!) by a whopping 38% faster, and therefore outperforms Power8 on a per core basis! Considering that SPARC M7 has an Oracle license core multiplier of .5x versus Power8 being a 1x multiplier, you'll need two times more Oracle licenses on Power8 to deliver same performance than SPARC M7. Poor Oracle is going to lose Database revenue stream selling on SPARC M7!
https://blogs.oracle.com/BestPerf/entry/20160317_sparc_t7_1_oltp
And if you're looking for fastest response times, SPARC M7 cant be beat either. Just look at the SPECjbb2015 Critical-OPs benchmark where its based on SLA response times. Theres no system or CPU that comes close to SPARC M7 even when calculated on performance/core. Not even Power8 which is 4.6x slower! https://blogs.oracle.com/BestPerf/entry/201511_specjbb2015_t7_1
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
