HP NC510F Service Manual

Retired
QuickSpecs
HP NC510F PCIe 10 Gigabit Server Adapter
HP NC510F PCIe 10 Gigabit Server Adapter
HP NC510F PCIe 10 Gigabit Server AdapterHP NC510F PCIe 10 Gigabit Server Adapter
Overview
The NC510F is an eight lane (x8) PCI Express (PCIe) 10 Gigabit Ethernet SR (10GBASE-SR fiber optic) network solution offering the highest bandwidth available in a ProLiant Ethernet adapter. The NC510F is ideal for a host of demanding applications meeting the needs of the next generation datacenter.
The NC510F ships with advanced server features that ProLiant customers have come to expect, such as support for TCP checksum and segmentation (LSO) offload capability, jumbo frames, IPv6, and more. Flexible to order and install - either standard or low profile slot capability with only one part number.
Models
Models
ModelsModels
NOTE:
NOTE:
A minimum of 1 gigabyte (1 GB) of server memory is required per each NC510F adapter.
NOTE:NOTE:
Kit Contents
Kit Contents
Kit ContentsKit Contents
The option kit contents consist of the following items:
DA - 12588 Worldwide — Version 18 — April 27, 2009
414126-B21
Page 1
Retired
QuickSpecs
Overview
NOTE:
NOTE:
NOTE: NOTE: specifications, please visit:
Standard Features
Standard Features
Standard FeaturesStandard Features
Server Support
Server Support
Server SupportServer Support
For a brief, printer friendly data sheet that describes this product and informs you of the essential capabilities and
http://h71028.www7.hp.com/ERC/downloads/4AA0-6064ENW.pdf
ProLiant DL (rack-optimized)
ProLiant DL (rack-optimized)
ProLiant DL (rack-optimized)ProLiant DL (rack-optimized)
HP ProLiant DL140 G2 HP ProLiant DL140 G3 HP ProLiant DL145 G2 HP ProLiant DL145 G3 HP ProLiant DL160 G5 HP ProLiant DL165 G5 HP ProLiant DL165 G5p HP ProLiant DL180 HP ProLiant DL180 G5 HP ProLiant DL185 G5 HP ProLiant DL360 G4p HP ProLiant DL360 G5 HP ProLiant DL365 HP ProLiant DL365 G5 HP ProLiant DL380 G4 HP ProLiant DL380 G5 HP ProLiant DL385 G2 HP ProLiant DL385 G5 HP ProLiant DL385 G5p HP ProLiant DL580 G3 HP ProLiant Dl580 G5 HP ProLiant DL585 G2 HP ProLiant DL585 G5 HP ProLiant DL785 G5
HP NC510F PCIe 10 Gigabit Server Adapter
HP NC510F PCIe 10 Gigabit Server Adapter
HP NC510F PCIe 10 Gigabit Server AdapterHP NC510F PCIe 10 Gigabit Server Adapter
.
:
ProLiant ML (expansion-optimized)
ProLiant ML (expansion-optimized)
ProLiant ML (expansion-optimized)ProLiant ML (expansion-optimized)
HP ProLiant ML350 G5 HP ProLiant ML370 G5 HP ProLiant ML570 G3 HP ProLiant ML570 G4
NOTE:
NOTE:
NOTE: NOTE:
Some servers may be discontinued.
DA - 12588 Worldwide — Version 18 — April 27, 2009
:
Page 2
Retired
QuickSpecs
Performance
Performance
Performance
PerformancePerformance
10 Gigabit Throughput
10 Gigabit Throughput
10 Gigabit Throughput10 Gigabit Throughput
TCP/IP Offload Engine
TCP/IP Offload Engine
TCP/IP Offload EngineTCP/IP Offload Engine
Accelerated iSCSI
Accelerated iSCSI
Accelerated iSCSIAccelerated iSCSI
The 10,000 Mbps Ethernet transfer rate delivers outstanding network performance that improves response time and removes bottlenecks required for the next generation datacenter. Ten Gigabit Ethernet bandwidth is ideal for high performance computing, database clusters, network attached storage, storage backups, grid systems, virtualization, server and I/O consolidation, fabric consolidation, and more.
TCP/IP Offload Engine (TOE) moves the processing of data in the TCP protocol stack from the server CPU to the network card, freeing CPU cycles for other duties. With TOE, network communications are improved, and server efficiency is increased. The TOE multifunction feature is planned for future, near­term release available for both Windows and Linux.
Internet Small Computer System Interface (iSCSI) allows the NC510F to multi-function both as a network adapter and as a storage host bus adapter, thereby providing access to storage boxes and to servers over a single connection. Accelerated iSCSI offloads the iSCSI software to the hardware for improved system performance. The accelerated iSCSI multifunction feature is planned for future, near-term release.
HP NC510F PCIe 10 Gigabit Server Adapter
HP NC510F PCIe 10 Gigabit Server Adapter
HP NC510F PCIe 10 Gigabit Server AdapterHP NC510F PCIe 10 Gigabit Server Adapter
RDMA
RDMA
RDMARDMA
PCI Express Interface
PCI Express Interface
PCI Express InterfacePCI Express Interface
Jumbo Frames
Jumbo Frames
Jumbo FramesJumbo Frames
TCP Stateless Offloading
TCP Stateless Offloading
TCP Stateless OffloadingTCP Stateless Offloading
Remote Direct Memory Access (RDMA) is a low-latency protocol providing improved throughput scaling and message latency between two RDMA-capable systems. RDMA allows data from the memory of one computer to communicate directly with the memory of another minimizing CPU and memory overhead. The RDMA multifunction feature is planned for future, near-term release.
HP was an early champion of PCI Express bus technology and has played a key role in the development and industry adoption of the PCI Express specification. The NC510F is designed with an eight lane (x8) PCI Express bus based on the PCIe v1.1 standard. The adapter is backward compatible with four lane (x4) PCI Express automatically auto-sensing between x8 and x4 slots.
Jumbo frames (also known as extended frames) permit a 8,000 byte transmission unit (MTU), which is five times the size of standard 1500 byte Ethernet frame. The NC510F supports jumbo frames as a way to achieve higher throughput and better CPU utilization. Jumbo frames are particularly useful for database transfers and tape backups.
For overall improved system response, the NC510F supports standard TCP/IP offloading techniques including:
TCP/IP checksum offload (TCO) moves the TCP and IP checksum offloading from the CPU to the network adapter. Large Send Offload (LSO) or TCP Segmentation Offload (TSO) allows the TCP segmentation to be handled by the adapter rather than the CPU.
Interrupt Coalescing
Interrupt Coalescing
Interrupt CoalescingInterrupt Coalescing
Interrupt coalescing (interrupt moderation) groups multiple packets, thereby reducing the number of interrupts sent to the host. This process optimizes host efficiency, leaving the CPU available for other duties.
DA - 12588 Worldwide — Version 18 — April 27, 2009
Page 3
Loading...
+ 5 hidden pages