Newsletters   Subscriptions  Forums  Store   Career  Media Kit  About Us  Contact  Search   Home 
tlb
Volume 2, Number 36 -- September 27, 2005

Dell Starts Peddling Dual-Core Paxville Xeon DPs in PowerEdges


by Timothy Prickett Morgan


While the dual-core "Paxville" Xeon DP processors, which were not even a planned product until fairly recently, are not quite ready to ship in volume, Dell has once again received the pole position in the Intel server processor race and is being allowed to sell PowerEdge servers using Intel's first dual-core Xeon DP ahead of all other server vendors. This is a benefit--and perhaps the main one--of being Intel's staunchest supporting in the chip war that is raging between itself and Advanced Micro Devices.

Back in July, when Intel starting rolling out the "Smithfield" Pentium D dual-core Pentium 4 processors, Dell also got first crack at putting these chips into its entry PowerEdge SC430 server, and then into the PowerEdge 830 tower and the 850 rack-mounted machine at the end of August. According to Bruce Kornfeld, director of worldwide enterprise systems marketing at Dell, customers who sign a non-disclosure agreement can get access to the feeds and speeds of the Dell PowerEdge boxes that use the Paxville DP processors and actually place orders for the machines even though the official general availability of the servers from Dell is not expected until next month.

Intel had slated a Paxville Xeon MP processor for servers early next year, but when it became apparent that AMD was going to make a lot of hay out of its significant lead in dual-core X64 processors, Intel quickly created a version of the Paxville that would fit in the Xeon DP thermal profile and also work with the existing "Lindenhurst" chipsets that the single-core "Irwindale" Xeon DPs use today. Not only has Intel created a Paxville Xeon DP chip, but it has accelerated the delivery of the Paxville Xeon MP processor into the end of this year to try to blunt AMD's advantages in the market as it awaits the delivery of the real dual-core Xeon DP processor, "Dempsey," and its so-called "Bensley" platform. Dell, being the sole tier one server supplier that is Intel-only from top to bottom, is always positioned to get in line first with new Intel technologies, and this is absolutely intentional.

Kornfeld said that Dell would be supporting the Paxville Xeon DP processors in its rack-mounted PowerEdge 1850 (two-socket, 1U rack), PowerEdge 2800 (two-socket, tower), and PowerEdge 2850 (two-socket, 2U rack) servers. The Paxville DPs will also be available in the PowerEdge 1855 two-socket blade servers that Dell announced a little less than a year ago, and it its Precision 470 and 670 two-socket workstations. Kornfeld said that customers have to flash the BIOS on these machines to have them be able to see the two cores on each chip, but that this BIOS update would be absolutely compatible with any software running on the PowerEdge machines and Precision workstations. Dell is offering Microsoft's Windows Server 2003 and Red Hat's Enterprise Linux 4 pre-installed at the factory on these servers. Prices for base configurations of these machines (which Dell would not specify) range from $2,448 to $2,779. (This information, in the absence of configuration data, is nearly meaningless.) The Paxville Xeon DPs run at 2.8 GHz and have 2 MB of cache per core.

As you might expect, Kornfeld changed the subject a number of times when asked about how the revamped PowerEdge servers would stack up against servers using the dual-core 2.4 GHz Opteron 280 processors that were also announced yesterday. He said that what Dell customers who have gone under NDA are concerned with is how much extra performance they can get in their existing machines at what price and at what incremental increase in heat. What Dell wants to focus on is the fact that customers can see as much as 53 percent more performance by going with dual-core Paxville Xeon DPs.


On a set of benchmarks that Intel ran back in June and that Dell is using in its promotional material, Intel compared a two-socket Paxville Xeon DP (that's four cores running at 2.8 GHz) compared to the same machine using single-core Irwindale Xeon DPs running at 3.6 GHz. Running Microsoft Exchange email messaging, the Paxville DP box showed an 18 percent performance increase, while the TPC-C online transaction processing benchmark test did a little better with a 37 percent performance boost. Running the SPECjbb2000 Java benchmark (which is a loose interpretation of the TPC-C workload as implemented in Java), the Paxville DP saw a 43 percent boost in performance. And in the SPECint_rate_base2000 benchmark, which measures raw integer number crunching performance, the Paxville DP showed 53 percent higher throughput than a single-core Xeon DP. These are some of the biggest performance numbers that Intel has brought to market in a long time.

That performance boost comes at a slight cost, though: slightly larger power supplies. In the PowerEdge 1850 with two single-core 3.6 GHz Xeon DPs, Dell used a 460 watt power supply and the machine had a SPECint_rate2000 performance of 38.1; moving to the dual-core Paxville Xeon DPs, which stepped down to 2.8 GHz, still required a 500 watt power supply, but the server had a SPECint_rate2000 rating of 58.5. That is an 8.7 percent increase in watts for a 53.5 percent increase in performance, or a 41.3 percent increase in performance per watt. A similar comparison with the PowerEdge 2850, which has a larger power supply to start with, yielded nearly identical performance and slightly better improvement in performance per watt. This is what Kornfeld says Dell's customers care about.

When Intel does get the Dempsey Xeon DPs and Bensley platforms out the door early next year, it seems like a safe bet that Dell will get to peddle these boxes first, too. And they will offer significantly better numbers than these last-minute Paxville Xeon DP boxes do. The question now is whether or not Intel and, by extension, Dell are shooting themselves in the foot. What if everyone with current Xeon DP boxes using the Lindenhurst chipset just decides to buy processor upgrades?

Sponsored By
LINUX NETWORX

Clusterworx® Whitepaper

High performance Linux clusters can consist of hundreds or thousands of individual components. Knowing the status of each CPU, memory, disk, fan, and other components is critical to ensure the system is running safely and effectively.

Likewise, managing the software components of a cluster can be difficult and time consuming for even the most seasoned administrator. Making sure each host's software stack is up to date and operating efficiently can consume much of an administrator's time. Reducing this time frees up system administrators to perform other tasks.

Though Linux clusters are robust and designed to provide good uptime, occasionally conditions lead to critical, unplanned downtime. Unnecessary downtime of a production cluster can delay a product's time to market or hinder critical research.

    Since most organizations can't afford these delays, it's important that a Linux cluster comes with a robust cluster monitoring tool that:
  • Provides essential monitoring data to make sure the system is operational.
  • Eliminates repetitive installation and configuration tasks to reduce
          periods of downtime.
  • Provides powerful features, but doesn't compromise on usability.
  • Automates problem discovery and recovery on would-be critical events.

This paper discusses the features and functions of Clusterworx® 3.2. It details how Clusterworx® provides the necessary power and flexibility to monitor over 120 system components from a single point of control. The paper also discusses how Clusterworx® reduces the time and resources spent administering the system by improving software maintenance procedures and automating repetitive tasks.

High Performance Monitoring

Each cluster node has its own processor, memory, disk, and network that need to be independently monitored. This means individual cluster systems can consist of hundreds or thousands of different components. The ability to monitor the status and performance of all system components in real time is critical to understanding the health of a system and to ensure it's running as efficiently as possible.

Because so many system components need to be monitored, one of the challenges of cluster management is to efficiently collect data and display system health status in an understandable format. For example, let's say a cluster system has 100 nodes and is running at 97 percent usage. It's very important to know whether 100 nodes are running at 97 percent usage or whether 97 nodes are running at 100 percent usage while three nodes are down.

Clusterworx® provides real-time analysis of over 120 essential system metrics from each node. Data is displayed in easy-to-read graphs, thumbnails, and value tables. Clusterworx® collects data from groups of nodes to spot anomalies, then drills down to single node view to investigate problems. This allows users to determine exactly what the problem is before taking corrective action.

Clusterworx® also tracks the power and health state of each node and displays its status using visual markers in a node tree view throughout the user interface. Power status shows whether the node is on, off, provisioning, or in an unknown state. The health state tracks informational or warning messages and critical errors. Health state messages are displayed in a message queue on the interface.

Clusterworx®'s comprehensive monitoring and easy-to-read charts and graphs allow users to quickly asses the state of each node and the overall system at a glance - while providing the necessary information to make informed decisions about the cluster system.

To read the rest of this whitepaper, please visit www.linuxnetworx.com


Editor: Timothy Prickett Morgan
Contributing Editors: Dan Burger, Joe Hertvik, Kevin Vandever,
Shannon O'Donnell, Victor Rozek, Hesh Wiener, Alex Woodie
Publisher and Advertising Director: Jenny Thomas
Advertising Sales Representative: Kim Reed
Contact the Editors: To contact anyone on the IT Jungle Team
Go to our contacts page and send us a message.


THIS ISSUE
SPONSORED BY:

Egenera
Linux Networx
California Digital
Roaring Penguin
Novell


The Linux Beacon

BACK ISSUES

TABLE OF
CONTENTS
AMD Cranks Up Dual-Core Opteron Clocks

Dell Starts Peddling Dual-Core Paxville Xeon DPs in PowerEdges

IDC Concurs that Q2 Was Pretty Good for Servers

Egenera Gets $300 Million Reseller Deal with Fujitsu-Siemens

But Wait, There's More


The Four Hundred
iSeries Execs Talk Up the Future of the Platform at COMMON

COMMON Sound Off: Frustration Level Is Down a Bit Among the Faithful

Oracle to Support IBM's WebSphere with Project Fusion Apps

Mad Dog 21/21: New Moth

The Windows Observer
Microsoft Reorganizes Ahead of Allchin's Retirement in 2006

Microsoft Refines Software Assurance for 2006

Softricity Streamlines Access to Desktop Apps with ZeroTouch

Egenera Gets $300 Million Reseller Deal with Fujitsu-Siemens

The Unix Guardian
UltraSparc-IV+ Chips Give Sun's Midrange Servers Twice the Oomph

Egenera Gets $300 Million Reseller Deal with Fujitsu-Siemens

IBM Shifts Its SOA Initiative Up Into High Gear

Notes/Domino 7 Brings New Collaboration Technology, Performance Gains


Copyright © 1996-2008 Guild Companies, Inc. All Rights Reserved.
Guild Companies, Inc. (formerly Midrange Server), 50 Park Terrace East, Suite 8F, New York, NY 10034
Privacy Statement