Posts Tagged ‘Flash Express’

IBM z13 Chip Optimized for Top Enterprise Performance

January 23, 2015

With the zEC12 IBM boasted of the fastest commercial chip in the industry. It is not making that boast with the z13. Instead, it claims a 40% total capacity improvement over the zEC12. The reason: IBM wants the z13 to excel at mobile, cloud, and analytics as well as fast extreme scale transaction processing. This kind of performance requires optimization up and down the stack; not just chip speed but I/O processing, memory access, instruction tweaks, and more.

 z13 mobile

Testing mobile transactions on the z13

This is not to suggest that the machine is not fast.  It is.  Timothy Prickett Morgan writing in his 400 blog notes that the z13 chip runs a 22 nm core at 5 GHz, half a GHz slower than the zEC12. The zEC12 processor, the one touted as the fastest commercial processor, was a 32nm core that clocked at 5.5 GHz.  Still, the z13 delivers about a 10 percent performance bump per core thanks, he writes, to other tweaks in the core design, such as better branch prediction and better pipelining in the core. The slightly slower clock speed reduces heat.

Up and down the stack IBM has been optimizing the z13 for maximum performance.

  • 2X performance boost for cryptographic coprocessors
  • 2X increase in channel speed
  • 2X increase in I/O bandwidth
  • 3X increase in memory capacity
  • 2X increase in cache and a new level of cache

At 5 GHz, the z13, given all the enhancements IBM has made, remains the fastest. According to IBM, it is the first system able to process 2.5 billion transactions a day, equivalent of 100 Cyber Mondays every day of the year.  Maybe even more importantly, z13 transactions are persistent, protected, and auditable from end-to-end, adding assurance as mobile transactions grow to an estimated 40 trillion (that’s trillion with a T) mobile transactions per day by 2025.

Given that mobile is shaping up to be the device of the future the z13 is the first system to make practical real-time encryption of all mobile transactions at any scale, notes IBM. Specifically, the z13 speeds real-time encryption of mobile transactions to help protect the transaction data and ensure response times consistent with a positive customer experience.  With mobile overall, the machine delivers up to 36% better response time, up to 61% better throughput, and up to 17% lower cost per mobile transaction. And IBM discounts transactions running on z/OS.

To boost security performance the machine benefits from 500 new patents including cryptographic encryption technologies that enable more security features for mobile initiated transactions. In general IBM has boosted the speed of encryption up to 2x over the zEC12 to help protect the privacy of data throughout its life cycle.

Combined with the machine’s embedded analytics it can provide real-time insights on all transactions. This capability helps enable an organization to run real-time fraud detection on 100 percent of its business transactions.  In terms of analytics, the machine deliver insights up to 17x faster at 13x better price performance than its competitors.

Further boosting performance is the increase of memory in the machine. For starters, the machine can handle up to 10 TB of memory onboard to help with z/OS and Linux workloads. To encourage organizations to take advantage of the extra memory IBM is discounting the cost of memory. Today memory runs $1500/GB but organizations can populate the z13 with new memory starting at $1000/GB. With various discounts you can get memory for as little as $200/GB.

So what will you do with a large amount of discounted memory? Start by running more applications in-memory to boost performance.  Do faster table scans in memory to speed response or avoid the need for I/O calls. Speed sorting and analytics by doing it in memory to enable faster, almost real-time decision making. Or you can run more Java without increasing paging and simplify the tuning of DB2, IMS and CICS. Experience 10x faster response time with Flash Express and a 37% increase in throughput compared to disk.

As noted above IBM optimized just about everything that can be optimized. It provides 320 separate channels dedicated just to drive I/O throughput as well as performance goodies only your geeks will appreciate like simultaneous multithreading (SMT), symmetric multiprocessing (SMP), and single instruction, multiple data (SIMD). Overall about 600 processors (in addition to your configurable cores) speed and streamline processes throughout the machine.

Mainframes have the fastest processors in the industry –none come close–and with the addition of more memory, faster I/O,  and capabilities like SMT and SIMD noted above, the z13 clearly is the fastest. For workloads that benefit from this kind of performance, the z13 is where they should run.

DancingDinosaur is Alan Radding, a long-time IT analyst and writer. You can follow DancingDinosaur on Twitter, @mainframeblog. Check out his other IT writing at Technologywriter.com and here.

zEnterprise EC12: the Next Hybrid Mainframe

August 30, 2012

On Tuesday, IBM launched the zEnterprise EC12 (zEC12), a machine it had been hinting at for months as zNext, the next hybrid mainframe. As you would expect from the latest release of the top-of-the-line mainframe, the zEC12 delivers faster speed and better price/performance. With a 5.5 GHz core processor, up from 5.2 GHz in the z196, and an increase in the number of cores per chip (from 4 to 6) IBM calculates it delivers 50% more total capacity in the same footprint. The vEC12 won’t come cheap but on a cost per MIPS basis it’s probably the best value around.

More than just performance, it adds two major new capabilities, IBM zAware and Flash Express, and a slew of other hardware and software optimizations. The two new features, IBM zAware and Flash Express, both promise to be useful, but neither is a game changer. IBM zAware is an analytics capability embedded in firmware. It is intended to monitor the entire zEnterprise system for the purpose of identifying problems before they impact operations.

Flash Express consists of a pair of memory cards installed in the zEC12; what amounts to a new tier of memory. Flash Express is designed to streamline memory paging when transitioning between workloads. It will moderate workload spikes and eliminate the need to page to disk, which should boost performance.

Unless you are finding it difficult to keep your z machines running or are experiencing paging problems these capabilities won’t be immediately helpful.  They really are intended for shops with the most demanding workloads and no margin for error. The zEC12 also continues IBM’s hybrid computing thrust by including the zBX and new capabilities from System Director to be delivered through Unified Resource Manager APIs. You’ll need a zBX mod 3 to connect to the zEC12.

This is a stunningly powerful machine, especially coming just 25 months after the z196 introduction. The zEC12 is intended for optimized corporate data serving. Its 101 configurable cores deliver a performance boost for all workloads. The zEC12 also comes with the usual array of assist processors, which are just configurable cores with the assist personality loaded on. Since they are EC12 cores, they bring a 20% MIPS price/performance boost.

The processor has been optimized for better software performance, particularly for Java, PL/1, and DB2 workloads.  As with the z196, it handles out of order instruction processing and multi-level branch prediction for complex workloads. The new machine’s larger L2, L3, and L4 caches deliver almost 2x more on the chip to speed data to the processor. In addition, Flash Express provides 1.6 TB of usable capacity (packaged in pairs for redundancy, 3.2 TB total).

IBM estimates up to a 45% improvement in Java workloads, up to a 27% improvement in CPU-intensive integer and floating point C/C++ applications, up to 30% improvement in throughput for DB2 for z/OS operational analytics, and more than 30% improvement in throughput for SAP workloads. IBM has, in effect, optimized the zEC12 from top to bottom of the stack. DB2 applications are certain to benefit as will WebSphere and SAP.

IBM characterizes zEC12 pricing as follows:

  • Hardware—20% MIPS price/performance improvement for standard engines and specialty engines , Flash Express runs $125,000 per pair of cards (3.2 TB)
  • Software—update pricing will provide 2%-7% MLC price/performance for flat-capacity upgrades from z196, and IFLs will maintain PVU rating of 120 for software  yet deliver more 20% MIPS
  • Maintenance—no less than 2% price performance improvement for standard MIPS and 20% on IFL MIPS

IBM is signaling price aggressiveness and flexibility to attract new shops to the mainframe and stimulate new workloads. The deeply discounted Solution Edition program will include the new machine. IBM also is offering financing with deferred payments through the end of the year in a coordinated effort to move these machines now.

As impressive as the zEC12 specifications and price/performance is DancingDinosaur is most impressed by the speed at which IBM delivered the machine. It broke with its with its historic 3-year release cycle to deliver this potent hybrid machine just two years after the z196 first introduced hybrid computing.


Follow

Get every new post delivered to your Inbox.

Join 795 other followers

%d bloggers like this: