The following registry keywords in Windows Server 2003 are no longer supported and are ignored in Windows Server 2008:
HKLM\System\CurrentControlSet\Services\Tcpip\Parameters
HKLM\system\CurrentControlSet\Services\Tcpip\Parameters
HKLM\system\CurrentControlSet\Services\Tcpip\Parameters
Network-Related Performance Counters
This section lists the counters that are relevant to managing network performance. Most of the counters are straightforward. We provide guidelines for the counters that typically require explanation.
IPv4
Datagrams received per second.
Datagrams sent per second.
Network Interface > [adapter name]
Bytes received per second.
Bytes sent per second.
Packets received per second.
Packets sent per second.
Output queue length.
This counter is the length of the output packet queue (in packets). If this is longer than 2, delays occur. You should find the bottleneck and eliminate it if you can. Because NDIS queues the requests, this length should always be 0.
Packets received errors.
This counter is the number of incoming packets that contain errors that prevent them from being deliverable to a higher-layer protocol. A zero value does not guarantee that there are no receive errors. The value is polled from the network driver, and it can be inaccurate.
Packets outgoing errors.
Processor
Percent of processor time.
Interrupts per second.
DPCs queued per second.
This counter is an average rate at which DPCs were added to the processor's DPC queue. Each processor has its own DPC queue. This counter measures the rate that DPCs are added to the queue, not the number of DPCs in the queue. It displays the difference between the values that were observed in the last two samples, divided by the duration of the sample interval.
TCPv4
Connection failures.
Segments sent per second.
Segments received per second.
Segments retransmitted per second.
Decisions about how to design or configure storage software and hardware usually consider performance. Performance is always degraded or improved because of trade-offs with other factors such as cost, reliability, availability, power, or ease of use. Trade-offs are made all along the way between application and disk media. File cache management, file system architecture, and volume management translate application calls into individual storage access requests. These requests traverse the storage driver stack and generate streams of commands that are presented to the disk storage subsystem. The sequence and quantity of calls, and the subsequent translation, can improve or degrade performance.
Figure 2 shows the storage architecture, which covers many components in the driver stack.
Figure 2. Storage Driver Stack
The layered driver model in Windows sacrifices some performance for maintainability and ease of use (in terms of incorporating drivers of varying types into the stack). The following sections discuss tuning guidelines for storage workloads.
| 