BladeCenter processor blades, I/O expansion adapters, and units

Abstract

This paper describes the electrical architecture and design of the IBM eServere BladeCentert processor blades, expansion blades, and input/output (I/O) expansion adapters and units. The processor blades are independent, general-purpose servers containing processors, chipsets, main memory, hard drives, network interface controllers, power input control circuitry, and local systems management. The blade architecture is robust and flexible enough to enable the design of general-purpose Intelprocessor-based two-way and four-way blades, IBM POWEReprocessor-based blades, blades based on AMD Opterone processors, and common expansion blades. In addition, the processor blades, expansion adapters, and units use a serializeddeserialized (SerDes) interface for the internal I/O fabrics, thus giving blades the flexibility to support virtually any I/O protocol that supports SerDes. Support for I/O fabrics beyond the base Gigabit Ethernet is accomplished via optional I/O expansion adapters for Fibre Channel, Myricom Myrinet t, InfiniBand t, or additional Gigabit Ethernet. Additional storage or peripheral component interface capability can be added to the processor blades via expansion blades.

Scale-up x Scale-out: A Case Study using Nutch/Lucene

“…The IBM BladeCenter [3], [5]solution (and similar systems from companies such as HP and Dell) represents the next step after rack-mounted clusters in scale-out systems for commercial computing. The blade servers [6] used in BladeCenter are similar in capability to the densest rack-mounted cluster servers: 4-processor configurations, 16-32 GiB of maximum memory, built-in Ethernet, and expansion cards for either Fiber Channel, Infiniband, Myrinet, or 10 Gbit/s Ethernet. Also offered are double-wide blades with up to 8-processor configurations and additional memory.…”

BladeCenter networking

“…The technology chosen for the I/O expansion adapter must match the technology in switch module bays 3 and 4. For details, see switch packaging, midplane interconnection, and blade I/O expansion adapter [10][11][12].…”

BladeCenter chassis management

“…Chassis hardware is fully redundant and comprises fully redundant MMs, redundant buses to each blade [9], I/O switch module [10], redundant buses on the midplane [11], and power module, tachometer control to each blower, and redundant Ethernet connections to all I/O switch modules. This hardware architecture enables duplicate MMs to provide highly available chassis management services.…”