Optimizer (Independent Channel) Mode; Section 8. Chipset; Overview / Description; Intel 5220 Chipset (Cade Named Tylersburg) I/O Hub (Ioh) - Dell POWEREDGE R610 Technical Manualbook

Dell server user manual
Hide thumbs Also See for POWEREDGE R610:
Table of Contents

Advertisement

Dell™ PowerEdge™ R610 Technical Guidebook
h. Optimizer (independent channel) mode
In this mode, all three channels are populated with identical memory modules. This mode permits a
larger total memory capacity but does not support SDDC with x8-based memory modules.
A minimal single-channel configuration of 1GB memory modules per processor is also supported in
this mode.

sectiOn 8. chiPset

a. Overview / Description
The PowerEdge R610 planar incorporated the Intel 5520 chipset (code named Tylersburg) for I/O and
processor interfacing. Tylersburg is designed to support Intel's 5500 series processors (code named
Nehalem-EP), QPI interconnect, DDR3 memory technology, and PCI Express Generation 2. The
Tylersburg chipset consists of the Tylersburg-36D IOH and ICH9.
The Intel 5520 chipset (code named Tylersburg) I/O Hub (IOH)
The planar uses the The Intel
5520 chipset (code named Tylersburg) I/O Hub (IOH)-36D IOH to provide
®
a link between the 5500 series 2S processor (Nehalem EP) and I/O components. The main components
of the IOH consist of two full-width QuickPath Interconnect links (one to each processor), 36 lanes of
PCI Express Gen2, a x4 Direct Media Interface (DMI), and an integrated IOxAPIC.

IOH QuickPath Interconnect (QPI)

The QuickPath Architecture consists of serial point-to-point interconnects for the processors and the
IOH. The PowerEdge R610 has a total of three QuickPath Interconnect (QPI) links: one link connecting
the processors and links connecting both processors with the IOH. Each link consists of 20 lanes
(full-width) in each direction with a link speed of up to 6.4 GT/s. An additional lane is reserved for a
forwarded clock. Data is sent over the QPI links as packets.
The QuickPath Architecture implemented in the IOH and CPUs features four layers. The Physical layer
consists of the actual connection between components. It supports Polarity Inversion and Lane Reversal
for optimizing component placement and routing. The Link layer is responsible for flow control and the
reliable transmission of data. The Routing layer is responsible for the routing of QPI data packets. Finally,
the Protocol layer is responsible for high-level protocol communications, including the implementation of
a MESIF (Modify, Exclusive, Shared, Invalid, Forward) cache coherence protocol.
30

Hide quick links:

Advertisement

Table of Contents
loading

Table of Contents