RISC-y Business – Building the Case for Next-Generation UNIX Systems
0 likes117 views
1. Oracle discussed continuing to invest in and develop Sun's SPARC and storage platforms but provided little detail on roadmaps. This was likely aimed at reassuring Sun customers to prevent jumping ship. 2. Intel launched the Itanium 9300 processor with 2x performance but the announcement seemed rushed, possibly to overshadow IBM, and partners gave a muted response. Intel's focus is keeping partners like HP happy who rely on Itanium. 3. IBM's POWER7 launch demonstrated confidence as the leader with 40% of the UNIX market. POWER7 provides significant performance gains that will maintain IBM's leadership position. IBM also emphasized POWER7's role in driving end-to-end system innovation for the
RISC-y Business – Building the Case for Next-Generation UNIX Systems
- 1. RISC-y Business – Building the Case for Next-Generation UNIX Systems By Charles King, Pund-IT, Inc. Recent IT industry events have created a field day for those who think IT news comes in “threes”: 1. During an event highlighting the conclusion of its acquisition of Sun Microsystems, Ora- cle discussed plans for the company’s hardware division and said it will invest in Sun’s Ul- traSPARC microprocessor and server platforms, X64-based systems and storage hardware. The company did not discuss the UltraSPARC roadmap in detail, though it did say it is plan- ning upgrades for systems based on both the SPARC T series and the SPARC64 chips made by Sun partner Fujitsu. However, there was no discussion of the status of UltraSparc-RK "Rock" processors and related 'Supernova" systems, which are rumored to have been dis- continued. Oracle co-president Charles Phillips did, however, comment on Oracle’s plans to build integrated offerings for online transaction processing (OLTP) and other database- intensive business applications. Only general guidance was offered regarding the availabil- ity of these new systems. 2. Intel introduced the Itanium processor 9300 series (codenamed "Tukwila"), which the company said delivers more than double the performance of its predecessor, boosts scal- ability and adds reliability features to the platform. The new chip offers twice as many cores as its predecessor (four versus two), eight threads per processor, and increased cache, interconnect bandwidth, memory bandwidth and memory capacity. Intel discussed plans for “at least” two more generations of Itanium with "Poulson," (scheduled for delivery in two years), adding an advanced multi-core architecture, instruction-level and hyper- threading enhancements and new reliability features. Future Itanium processors in develop- ment are being designed for socket and binary compatibility with Intel Itanium 9300 proc- essor-based systems and software. The company expects OEM systems based on the new processors to ship within 90 days. 3. IBM’s Power Systems group launched its next-generation POWER7 microprocessor, along with a quartet of business class, enterprise class, high-end and high-performance servers featuring the new chips. According to IBM, POWER7 systems can deliver four times the per- formance and four times the virtualization capabilities and are three to four times more en- ergy efficient than Power6+ servers, but will sell for the same price as those solutions. POWER7 processors can now run up to 32 simultaneous tasks — with eight cores and four threads per core. The chips also support more memory and throughput and offer new fea- tures, including TurboCore, Intelligent Threads and Active Memory Expansion. These fea- tures will make POWER7 valuable for demanding enterprise applications, and will allow them to provide the basis of workload-optimized IBM solutions like last year’s Smart Ana- lytics Systems. In addition, POWER7 systems’ robust performance and flexibility make them appropriate for increasingly complex Smarter Planet workloads and processes. IBM’s new POWER7 systems will begin shipping between February 19th and March 16th. Ostensibly, these events all focused on the three vendors’ latest/greatest enterprise micro- processor architectures (RISC for IBM Power and Sun SPARC, EPIC for Intel’s Itanium) for supporting enterprise-class UNIX and Linux workloads. But what was really going on be- neath the surface of these announcements? In short, we suggest that the trio offers excel- 5
- 2. lent examples of vendors engaged in customer preservation (Oracle), partner management (Intel) and forward thinking (IBM). Before we examine the announcements, first consider the general state of the UNIX market. Once considered a nearly unassailable bastion of enterprise-class computing (only main- frame systems enjoy a higher status), UNIX solutions have been under increasing pressure from below due to the expanding capabilities and popularity of servers leveraging x86/64 Intel Xeon and AMD Opteron processors. That situation is likely to become increasingly fraught as higher-performance processors, including Intel’s Nehalem EX, come to market. How have vendors responded? Not surprisingly, by emphasizing and expanding their plat- form’s already robust RAS (reliability availability, security) features, as well as virtualization and other capabilities that even the best x86 solutions have trouble meeting, let alone sur- passing. While such strategies have tended to force UNIX solutions into higher and higher niches, enhanced performance has also allowed vendors to explore, develop and inhabit entirely new markets, including workload optimized systems Oracle SPARC So what do these announcements say about the status of these vendors? Perhaps more im- portantly, did any of the three say anything appreciably new or unexpected? In the case of Oracle/Sun, not much. The five-hour marathon webcast was longer on time than on salient details, particularly regarding Sun’s UltraSPARC server business. Yes, Oracle will continue selling and developing SPARC systems, including workload-optimized server/appliances. That doesn’t show much in the way of server strategy leadership — Oracle has been ship- ping similar systems based on HP hardware for months and IBM’s workload-optimized sys- tems are arguably more sophisticated in terms of targeting specific strategic markets. More important was what was left largely unsaid: How much time, effort and money Oracle will willingly invest in resurrecting a SPARC platform that has fallen on particularly hard times. That lack of transparency doesn’t matter if the essential purpose of the discussion was what we expect — to reassure Sun clients enough to keep most from jumping ship to IBM and HP. Without SPARC customers’ good will, the platform will continue its steady mar- ketplace decline, sapping any enthusiasm Oracle has for continuing the platform’s develop- ment. What should customers do? Keep very close watch on any changes Oracle makes in its server strategy or to the rather slender roadmap it presented during the launch. Intel Itanium The timing of Intel’s announcement says almost as much about the state of Itanium as did Tukwila’s technical details. Coming on the same day and nearly at the same time as the POWER7 launch, Intel appeared to be looking to overshadow IBM as much as trying to es- tablish thought or technology leadership. Some may see this as reading too much into press release minutia, but some curious issues reinforced our view, such as the low key re- sponses of Itanium OEMs. More significantly, the considerable three-month lag time be- tween Intel’s announcement and the likely appearance of 9300 systems suggests the launch may have been a bit improvised or premature. So what was point? First and fore- most, to keep partners, especially HP, happy. Since HP and other Itanium OEMs depend en- tirely on Intel for microprocessor development, it behooves the company to be aggressive in offering those partners good news to take to market. 6
- 3. Intel certainly deserves credit for delivering a 2X bump in performance, especially since less than a year ago the company decided to incorporate industry-standard DDR3 memory into Tukwila rather than the fully buffered DIMM memory of previous Itanium chips. De- spite that, it seems worth considering the platform’s longer-term prospects. At this point, HP represents some 85% of the Itanium market — a massive position, especially for a mi- croprocessor intended to be the Industry Standard platform for 64-bit enterprise comput- ing. The fact is that Itanium is feeling the same upward pressure from x86/64-based serv- ers as other UNIX platforms. We are not as pessimistic about Itanium as some assessments, including our colleague Joe Clabby’s in this issue of the Review. In short, we expect Itanium development will continue as long as HP’s Integrity solutions are profitable. But if HP ever begins porting HP-UX or its other legacy platforms to Intel Xeon, it will likely mark the offi- cial beginning of Itanium’s end. IBM POWER7 By comparison, IBM’s POWER7 announcement demonstrated the confidence of a vendor in firm control and thinking far ahead of its peers, and with good reason. Less than a decade ago, IBM trailed both Sun and HP in UNIX system sales. Today, it leads both those rivals with some 40% of the $16 billion UNIX market. As Itanium and SPARC development hit pot- holes, snapped axels and headed into the ditch, IBM’s Power efforts stayed remarkably on track and true to the company’s long term roadmap. POWER7’s significant performance im- provements mean that it is unlikely that any competing platforms or systems will pressure, let alone alter, IBM’s current UNIX performance leadership position where the company’s Power6/6+ systems own over a hundred #1 industry benchmarks. The POWER7 announcement also demonstrated an interesting development in systems strategy. Rather than specifically highlighting next-generation microprocessor improve- ments, IBM instead focused on the role POWER7 plays in driving end-to-end system innova- tion. The result? The company made extremely strong cases both for POWER7 delivering significant benefits to traditional business applications and for providing the engine driv- ing next-generation Smarter Planet workloads and processes. How this will play exactly in the market is unclear but as systems become ever more complex, it seems entirely possi- ble that microprocessors will increasingly take a back seat to overall server evolution. If that is the case, it will require Oracle to further ramp up its SPARC efforts and Intel to tighten development and messaging with its OEM partners. Overall, the events of the past week or two portend an increasingly interesting future for both RISC- and EPIC-based systems. Far from the inevitable death spiral some have prog- nosticated for UNIX servers, both IBM and Intel’s efforts suggest that innovation and com- petition remain alive and well. Even the uncertainty around Oracle’s intentions for SPARC is better than the Slough of Despond in which Sun previously wallowed. But that said, we also believe that IBM’s POWER7 represents the current bar by which upcoming Itanium 9300 and future SPARC systems will be judged. If Oracle, HP and others will not or cannot step up their games, the enterprise server market will likely be a very different place in three to five years than it is today. © 2009 Pund-IT, Inc. All rights reserved. About Pund-IT, Inc. Pund-IT emphasizes understanding technology and product evolution and interpreting the effects these changes will have on business customers and the greater IT marketplace. 7