Feeds

HP upgrades NonStop Servers

We don't want no stinking faults

  • alert
  • submit to reddit

Internet Security Threat Report 2014

ComputerWire: IT Industry Intelligence

Hewlett Packard Co made good on its promises and those made by the executives in charge of the former Tandem Computers line of fault tolerant machines to keep upgrading and improving these machines, which are a major backbone in the financial services industry and which are increasingly used for high-end e-business applications where uptime is vital,

Timothy Prickett Morgan writes

.

The Tandem line used to bear the Himalaya name at Compaq Computer Corp, which bought Tandem back in 1997 as it sought a foothold in the enterprise server market. It's is now called the NonStop line within the Business Critical Systems unit of HP, which acquired Compaq two months ago (as we all know).

The NonStop server line was refreshed the last time in May 2000, when the high-end Himalaya S74000s, which can have from two to 16 300MHz MIPS R12000 processors in a single node, the S7400 Himalaya servers, which had from two to sixteen of the 250MHz R10000 processors, and the S740 and S74 Himalaya servers, which came with two 300MHz R12000 or two 250MHz R10000 processors, respectively, were all announced. The latter S740 and S74 models were intended as development machines or nodes in distributed Himalaya networks and could not be upgraded to the larger S74000 and S7400 machines.

At yesterday's announcement, the underlying NonStop hardware was refreshed with the MIPS R14000 processor, which HP is using at 500MHz and 550MHz clock speeds in the new NonStop S76000 and NonStop S86000 servers. The MIPS processors not only run at higher clock speeds, but the memory controller in the NonStop machine has been tweaked and its speed has been goosed, as has the bandwidth on the NonStop system bus. Main memory in the NonStop nodes has also been quadrupled, with capacities per node of 1GB, 2GB, 4GB, or 16GB available.

The NonStop S76000 offers from two to 16 of the 500MHz R14000 processors, each equipped with 4MB of L2 cache running at 250MHz. The system bus speed on the S76000s runs at 100MHz, twice the speed of the bus in the S74000s it replaces. The resulting machine offers about 1.5 times the performance of the S74000 on the OLTP workloads typified in the Himalaya installed base. The top-end NonStop S86000 uses the 550MHz R14000 processor, which is equipped with an 8MB L2 cache running at 275MHz; the system bus on this machine runs a little faster than the S76000 at 110MHz. When you add it all up, this server node, which also offers from two to 16 processors per node, has about 1.9 times the performance of an equivalently configured S74000 server node. HP will also offer S760 and S76 development and network node machines to replace the S740 and S74 nodes. The company has also announced a new 18GB, 15K RPM disk drive, which is 25% faster than the existing disk drives used in the NonStop product line. The ServerNet internetworking uses to connect NonStop nodes has also been enhanced to balance against the processor performance enhancements and the I/O requirements that will necessarily follow from their use. The NASDAQ stock market in the U.S is one of HP's highest-profile NonStop customers and is not coincidentally one of the early adopters of the new NonStop machines. NASDAQ has been using the new gear in production for more than a week, in fact.

Yesterday's NonStop announcements were not only about hardware, of course. The real benefit and differentiator of Tandem machines has always been software. The NonStop SQL/MX 1.5 release of the fault tolerant database that is integral to the NonStop line has been tweaked to provide performance enhancements and sophisticated query functions. The NonStop operating system has also been fitted with extensible compilers and new optimizers, and the integrated Web server has been beefed up with the requisite alphabet soup of SOAP, UDDI, XML, and XSLT support to participate in Web services. HP has also improved Java support with Java 3 (J2SE JDK 1.3.1) and with Java servlets (JSP 1.1 and Servlet 2.2). Java performance on the NonStop servers has increased between 35% because of coding changes, and when coupled with the new hardware, customers who bought S74000 servers in 2000 could see twice the Java performance by moving to S76000 server nodes and over 2.5 times the performance by moving to S86000 server nodes. Further on the software front, HP has also signed up SeeBeyond and Tibco as middleware partners on the NonStop platform for enterprise application integration.

To some HP employees, the NonStop announcement was a bit of a homecoming. Tandem founder James Treybig, who is no longer part of the company, left HP three decades ago to start Tandem. Scott Stallard, vice president and general manager of HP's Business Critical Systems unit, spent 27 years at HP and remembers when Treybig left to found Tandem. "I wish we would have thought ahead and kept him at HP," he said at the NonStop announcement yesterday. "We view this as a family reunion. NonStop is an integral part of our business, not just something off to the side."

Pauline Nist, vice president and general manager of the NonStop line, said that the platform was attracting new customers, in part because of the marketing drive behind the Zero-Latency Enterprise (ZLE) solutions for the telecommunications and financial services industries. She said that in the last four quarters, 20% of revenue in the NonStop product line was to new customers, and that in the fourth quarter, new customers accounted for 24% of revenue. Exactly how much money this is remains unclear, but the NonStop servers are widely believed to account for more than $1bn in sales annually, with services revenues on top of that. Nist said yesterday that NonStop revenues increased 19 percent from the third quarter of 2001 to the fourth quarter, and that sales were up 4% for all of 2001.

HP said that it was on track to offer another MIPS processor upgrade in 2003, and would move the NonStop platform to the Intel Corp 64-bit Itanium processors in 2004.

© ComputerWire

Top 5 reasons to deploy VMware with Tegile

More from The Register

next story
Docker's app containers are coming to Windows Server, says Microsoft
MS chases app deployment speeds already enjoyed by Linux devs
'Hmm, why CAN'T I run a water pipe through that rack of media servers?'
Leaving Las Vegas for Armenia kludging and Dubai dune bashing
SDI wars: WTF is software defined infrastructure?
This time we play for ALL the marbles
'Urika': Cray unveils new 1,500-core big data crunching monster
6TB of DRAM, 38TB of SSD flash and 120TB of disk storage
Facebook slurps 'paste sites' for STOLEN passwords, sprinkles on hash and salt
Zuck's ad empire DOESN'T see details in plain text. Phew!
Windows 10: Forget Cloudobile, put Security and Privacy First
But - dammit - It would be insane to say 'don't collect, because NSA'
Oracle hires former SAP exec for cloudy push
'We know Larry said cloud was gibberish, and insane, and idiotic, but...'
Symantec backs out of Backup Exec: Plans to can appliance in Jan
Will still provide support to existing customers
prev story

Whitepapers

Forging a new future with identity relationship management
Learn about ForgeRock's next generation IRM platform and how it is designed to empower CEOS's and enterprises to engage with consumers.
Win a year’s supply of chocolate
There is no techie angle to this competition so we're not going to pretend there is, but everyone loves chocolate so who cares.
Why cloud backup?
Combining the latest advancements in disk-based backup with secure, integrated, cloud technologies offer organizations fast and assured recovery of their critical enterprise data.
High Performance for All
While HPC is not new, it has traditionally been seen as a specialist area – is it now geared up to meet more mainstream requirements?
Saudi Petroleum chooses Tegile storage solution
A storage solution that addresses company growth and performance for business-critical applications of caseware archive and search along with other key operational systems.