Multi core systems

The thinner the die the less power is required to move electrons around it, so the more efficient the processor should be and the less excess heat is likely to be generated. When Intel moved from 130nm to 90nm dies, for example, the power requirements for some of its chips actually rose. Likewise it had further problems when it moved to the 65nm silicon from which its Xeon server chips are now made.

In particular you should look for low TDP (Thermal Design Power) ratings which, although not quite a performance per watt measure, will at least tell you whether a processor has been designed with power efficiency in mind.

Multi-core ought to be good news when it comes to efficiency. It enables, for example, one 2-way quad-core server (effectively 8 processor cores) to replace four 2-way servers fitted with old-fashioned single-core chips. This means just one power supply and one motherboard generating heat instead of four. There’s also just one system disk spinning away, one set of cooling fans and so on — leading to quite clear and measurable gains in terms of reduced power and cooling requirements. But equally, the introduction of multi-core processors means more processing power in less space. This is where your blade chassis comes. Although rack-optimized servers are still being deployed, for maximum processor density and the ultimate in potential efficiency, blade servers are the way forward.

Two major players in Blade chassis segments are IBM and HP

HP C-Class Chassis:

About the only difference between the c3000 and HP’s high-end BladeSystem c7000 is that the c3000 has a smaller chassis that holds fewer blades and so costs less. The c7000 takes a maximum of 16 blades, while the c3000 is limited to eight; however, the blades that go inside both enclosures are exactly the same.

Entry level c3000 comes with 2 PSU’s and 4 fans, which are sufficient for running ONLY 2 blades and enclosure link module, used to connect several c3000 systems.

c7000 enclosure

10U size with 8- full height blades or 16 half-height blades

Interconnect bays

8 bays; up to 4 redundant I/O fabrics

2 onboard administrator

465/685 are all Dual core –i.e. AMD

460/680 are either dual/quad core i.e. Intel

IBM Blade center:

IBM’s blade server product line is based around the 14-slot BladeCenter chassis, which it launched in 2002. Last year, IBM launched an upgrade, known as BladeCenter H, which added support for high-speed connectivity such as InfiniBand, along with larger power supplies and cooling systems. It also makes a ruggedised version, the BladeCenter T, which it sells largely to telcos and military establishments, in both standard and high-bandwidth versions.

How much space is left in C-Drive?

fsutil volume diskfree c:
Total # of free bytes        : 11735273472
Total # of bytes             : 36405055488
Total # of avail free bytes  : 11735273472

 


How many drives are present on your OS?

fsutil fsinfo drives

Drives: C:\ D:\ E:\ F:\ G:\

 


 

image

 

image

image

Screen Shot from Process Explorer

image

IRP :- I/O request packets

C:\windows\system32\drivers\partmgr.sys:
        Verified:       Signed
        Signing date:   14:28 04/08/2004
        Publisher:      Microsoft Corporation
        Description:    Partition Manager
        Product:        Microsoft« Windows« Operating System
        Version:        5.1.2600.0
        File version:   5.1.2600.0 (XPClient.010817-1148)

 

Ntldr resides on the system volume; the boot-sector code on the system volume executes Ntldr. Ntldr reads the Boot.ini file from the system volume and presents the computer’s boot choices to the user. The partition names that Boot.ini designates are in the form multi(0)disk(0)rdisk(0)partition(1).

CPU column of task manager,corresponds to spending time in an execution state during the last Task Manager snapshot.

As mentioned previously, the thread is the dispatchable unit in Windows 2000, not the process. Every thread represents an instruction stream that is eligible to execute on the processor.

When you boot from the Windows CD or the boot disks, by the time the system gives you the choice to install or repair Windows, the CD has booted a copy of the Windows kernel, including all necessary supporting device drivers (for example, NTFS or FAT drivers, SCSI drivers, a video driver).

Dynamic disks are partitioned using Logical Disk Manager (LDM) partitioning.

 

Mirrored volumes are the only multipartition volume type supported for system and boot volumes. The reason for this is that the Windows boot code, including the MBR code and Ntldr, don’t have the sophistication required to understand multipartition volumes—mirrored volumes are the exception because the boot code treats them as simple volumes, reading from the half of the mirror marked as the boot or system drive in the MBR-style partition table. Because the boot code doesn’t modify the disk, it can safely ignore the other half of the mirror.

http://support.microsoft.com/kb/888732

  • image

image

The combination of physical RAM and pagefile is called Virtual Memory. Virtual Memory is managed by Virtual Memory Manager. Physical memory(RAM) can be accessed at expoentially high speed than disk. So everytime server has to move data between physical disk and Memory, disk will introduced significant delay. Excessive and consistent paging activity is called thrashing.

Pagefile could be set in several ways, if you let Windows let manage PageFile size, it will create pagefile size (PhysicalRAM+1) MB. This PageFile size is required to catch memory dump when server create DUMP. Create seperate pagefile on seperate partition to improve performance. Number of pagefile you can create is 16 and Maximum size is 4 GB. So totak 16 x 4 =64GB page file size can be created and OS treats this file as one file. When pagefile is distrubuted across volume, all volumes have same characteristic like speed,size.Pagefile size should be kept on dedicated drive if we are looking for optimizing it’s use. RAID0/RAID1 might be good option but RAID5 is not recommended. Paging is write operation and may not suite characteristic of RAID5. If Page file optimization is required try not to put on system drive. If you need to catch memory, you have no option but to create it on System drive. Make sure you put on same drive as C drive and not same volume but different drive. Set Minimum and Maximum page file size to be same to avoid any overhead for dynamically increasing page file size. Best practise to create page file is to format volume and create pagefile before placing any data on that volume, this ensures you get full contiguous space.Also to note first partition on physical disk is closer to tip of this disk, where disk speed is highest and performance is best. How do you decide if you need more ram based on pagefile. Check pagefile: %usage Max in perfmon, if the usage is more than 10% it is time to think.

File system cache

ScreenShot063

File system cache is area on physical memory (RAM) for dynamically store accessing recently data of IO Subsystem, which includes Disks,Network interface and Networks. More is this space less times IO subsystem is accessed considerably increasing performance.

Maximize data throughput for file sharing: is generally recommended for file server and servers above 2GB RAM

Maximize data throughput for network applications: is using for application server and one who have internal memory optimization techinques.

Maximum values are put down above. SQL and Exchange server Maximize data throughput for network applications should be selected. Since they manage their internal system cache.

In a pure iSCSI clustering environment, you must set the Cluster service to depend on all the following services:

• MSiSCSI

• Clusnet

• Netman

• Rpcss

• W32time

 

sc config clussvc depend= clusnet/netman/Rpcss/w32time/msiscsi

 

More information is available KB 883397

 

Next Page »