Oracle's Sparc T5 Chip Evidently Pushed Back to 2013 98
Mark Hachman writes in Slash Datacenter that the Sparc T5 chip Oracle announced earlier this year apparently won't be ready until sometime in 2013. John Fowler, executive vice president, Systems, Oracle, presented at Oracle Open World a chart outlining highlights of Oracle's plans for the future.
"But Fowler also skipped over some bad news: an apparent delay for the Sparc T5. A year ago, Oracle’s Sun division announced the Sparc T4—and according to Fowler, Oracle chief Larry Ellison set a very high bar for the next iteration: double the performance while maintaining app compatibility on an annual basis. Apparently, that didn’t quite happen with the T5; Oracle had the opportunity to announce a T5-based server, and didn’t. That’s a bit of bad news for the Sun design team, which already had to watch Intel’s Xeon chief, Diane Bryant, give the preceding keynote. ... As detailed at this year’s Hot Chips conference, the T5 combines 16 CPU cores running at 3.6 GHz on a 28-nm manufacturing process. Continuing the trend of hardware acceleration of specific functions, Sun executives claimed the chip would lead in on-chip encryption acceleration, with support for asymmetric (public key) encryption, symmetric encryption, hashing up to SHA-512, plus a hardware random number generator."
Comment removed (Score:5, Informative)
Re:hardware, not software Re:Oracle? SPARC? (Score:5, Informative)
Sounds like SPARC has really gone nuts on the threading and core count. 16 cores with 8 threads each (per proc) makes for 128 threads per proc, which far outshines Intel and AMD.
Re:hardware, not software Re:Oracle? SPARC? (Score:5, Informative)
But since I haven't look at the non X86 chips in awhile my question is...what advantage do they give over the AMD and Intel X86 cores? i mean I could understand using MIPS, SPARC, POWER back in the day because X86 was slower at certain tasks but now the amount of IPC on the Intel side is just nuts and AMD is going with an insane numbers of cores for cheap...so what's the selling point?
I mean I can understand those that already have significant resources tied up in SPARC as it'd be cheaper to stay with what they have than to switch, but what do they use as selling points to get new customers?
For certain workloads it really shines. A couple of years ago, web servers was given as a prime example. Tons and tons of threads, none of them that powerful, but good for workloads where you have lots and lots of small-tasks running in parallel.
If you don't have that kind of workload, obviously it isn't attractive. But if you have the right kind of workload, you should be interested.
Re:hardware, not software Re:Oracle? SPARC? (Score:5, Informative)
Oracle SGBDR and options licences price is the key to understand why SPARC still matter.
For the same processing power, T4 cpu need less core than any x86 concurrent.
A client using spatial, partitionning and olap will need to rack up 66k$ per core (list price). Note that the licence requiered for a single core is more expensive than the server.
If you add that you only need to pay the licences you use on solaris thanks to the containers, and you can easyly imagine that a T4 solution usually is 2 times cheaper than a x86 solution.
Do I need to add the greater MTBF of these T4 machines or that it is the oracle developpement plateform of choice (thuss receive patches the first, windows is a 3rd cityzen in that context) ?