Thursday, February 10, 2011

Storage Replication


Storage Replication

Array-Based Replication
Server-Based Replication
SAN-Based Replication: A New Alternative

SRDF
SRDF (Symmetrix Remote Data Facility) is a family of EMC products that facilitates the data replication from one Symmetrix storage array to another through a Storage Area Network or IP network.

SRDF logically pairs a device or a group of devices from each array and replicates data from one to the other synchronously or asynchronously. An established pair of devices can be split, so that separate hosts can access the same data independently (maybe for backup), and then resynchronised.

In synchronous mode (SRDF/S), the primary array waits until the secondary array has acknowledged each write before the next write is accepted, ensuring that the replicated copy of the data is always as current as the primary. However, the latency due to propagation increases significantly with distance.

Asynchronous SRDF (SRDF/A) transfers changes to the secondary array in units called delta sets, which are transferred at defined intervals. Although the remote copy of the data will never be as current as the primary copy, this method can replicate data over considerable distances and with reduced bandwidth requirements and mimimal impact on host performance.

NetBackup Licensing


NetBackup licensing
Each server that you backup over the LAN needs a Standard Client license.  Each server backed up over the SAN needs an enterprise client license.  Each virtual machine backing up as a regular client will require a Standard client license.  Each virtual machine that will leverage Netbackup VCB integration, will require an Enterprise client license.

Each robotic tape drive will need a library based tape drive license.

The Netbackup server and NetBackup enterprise server license does not include any robotic tape drive licenses.

Application and Database licenses are licensed per database and per system, so you will need 2 Application and DataBase Pack Licenses to perform hot backups for both Oracle and Sybase.  Since both databases are running on the same OS, there is no need to purchase an additional client license

Storage Tiers


Storage Tier
0 - Special Functionality
1 - Enterprise (15,000 rpm)
2 - Modular (10,000 rpm)
3 - General Purpose (7,200 rpm SATA)
Connectivity Tier :
A - Fibre Attached
B - iSCSI Attached
C - NAS

Regular Charge Includes:
Storage Frames
Disk capacity,
Switch fabric and related fibre cabling.
Storage capacity for Unix, Linux and Windows server systems
RAID 5 based storage.
Data Centre costs
Third party premium maintenance
Depreciation of storage infrastructure

Regular Charge DOES NOT Include:
Server based licenses for SAN attachment (mutli-pathing software or volume manager software)
Host Bus Adaptors for connection of fibre
Fibre cabling from the customer server up to the storage fabric switch
Replication software (available as Tier 0)
Bandwidth and protocol convertors for cross site replication
Effort required to implement project requirements
Effort to administer and manage the SAN solution

C N A, Converged Network Adapters


Fibre Channel over Ethernet (FCoE) is an encapsulation of Fibre Channel frames over Ethernet networks. This allows Fibre Channel to use 10 Gigabit Ethernet networks (or higher speeds) while preserving the Fibre Channel protocol.
Many data centers use Ethernet for TCP/IP networks and Fibre Channel for storage area networks (SANs). With FCoE, Fibre Channel becomes another network protocol running on Ethernet, alongside traditional Internet Protocol (IP) traffic. FCoE operates directly above Ethernet in the network protocol stack, in contrast to iSCSI which runs on top of TCP and IP. As a consequence, FCoE is not routable at the IP layer, and will not work across routed IP networks.
Computers connect to FCoE with Converged Network Adapters (CNAs), which contain both Fibre Channel Host Bus Adapter (HBA) and Ethernet Network Interface Card (NIC) functionality on the same adapter card. CNAs have one or more physical Ethernet ports. FCoE encapsulation can be done in software with a conventional Ethernet network interface card, however FCoE CNAs offload (from the CPU) the low level frame processing and SCSI protocol functions traditionally performed by Fibre Channel host bus adapters.

Tuesday, February 1, 2011

Intel Virtualization Technolog


Intel Virtualization Technology (VT). Formerly known as Vanderpool, this technology enables a CPU to act as if it were several CPUs working in parallel, in order to enable several operating systems to run at the same time in the same machine. In this tutorial we will explain everything you need to know about this new technology.

You may confuse virtualization with multitasking or even with Hyper-Threading. On multitasking, there is a single operating system and several programs running in parallel.  On virtualization, you can have several operating systems running in parallel, each one with several programs running. Each operating system runs on a “virtual CPU” or “virtual machine”. And Hyper-Threading simulates two CPUs where there is just one physical CPU for balancing performance using SMP (Symmetric Multi Processing), and these two CPUs cannot be used separately.

Of course if a CPU has both Hyper-Threading and Virtualization Technology each virtual CPU will appear to the operating system as if two CPUs are available on the system for symmetric multiprocessing.

If you pay close attention, Virtualization Technology uses the same idea of Virtual 8086 (V86) mode, which is available since 386’s. With V86 mode you can create several virtual 8086 machines to run DOS-based programs in parallel. With VT you can create several “complete” virtual machines to run full operating systems in parallel.

CPUs with Virtualization Technology have some new instructions to control virtualization. With them, controlling software (called VMM, Virtual Machine Monitor) can be simpler, thus improving performance compared to software-only solutions.

How It Works ?
Processors with Virtualization Technology have an extra instruction set called Virtual Machine Extensions or VMX. VMX brings 10 new virtualization-specific instructions to the CPU: VMPTRLD, VMPTRST, VMCLEAR, VMREAD, VMWRITE, VMCALL, VMLAUCH, VMRESUME, VMXOFF and VMXON.

There are two modes to run under virtualization: root operation and non-root operation. Usually only the virtualization controlling software, called Virtual Machine Monitor (VMM), runs under root operation, while operating systems running on top of the virtual machines run under non-root operation. Software running on top of virtual machines is also called “guest software”.

To enter virtualization mode, the software should execute the VMXON instruction and then call the VMM software. Then VMM software can enter each virtual machine using the VMLAUNCH instruction, and exit it by using the VMRESUME. If VMM wants to shutdown and exit virtualization mode, it executes the VMXOFF instruction.

S C S I


SCSI
Small Computer System Interface, or SCSI (pronounced scuzzy[1]), is a set of standards for physically connecting and transferring data between computers and peripheral devices. The SCSI standards define commands, protocols, and electrical and optical interfaces. SCSI is most commonly used for hard disks and tape drives, but it can connect a wide range of other devices, including scanners and CD drives. The SCSI standard defines command sets for specific peripheral device types; the presence of "unknown" as one of these types means that in theory it can be used as an interface to almost any device, but the standard is highly pragmatic and addressed toward commercial requirements.
SCSI is an intelligent, peripheral, buffered, peer to peer interface. It hides the complexity of physical format. Every device attaches to the SCSI bus in a similar manner. Up to 8 or 16 devices can be attached to a single bus. There can be any number of hosts and peripheral devices but there should be at least one host. SCSI uses hand shake signals between devices, SCSI-1, SCSI-2 have the option of parity error checking. Starting with SCSI-U160 (part of SCSI-3) all commands and data are error checked by a CRC32 checksum. The SCSI protocol defines communication from host to host, host to a peripheral device, peripheral device to a peripheral device. However most peripheral devices are exclusively SCSI targets, incapable of acting as SCSI initiators—unable to initiate SCSI transactions themselves. Therefore peripheral-to-peripheral communications are uncommon, but possible in most SCSI applications. The Symbios Logic 53C810 chip is an example of a PCI host interface that can act as a SCSI target.

SAS
Serial Attached SCSI (SAS) is a computer bus used to move data to and from computer storage devices such as hard drives and tape drives. SAS depends on a point-to-point serial protocol that replaces the parallel SCSI bus technology that first appeared in the mid 1980s in data centers and workstations, and it uses the standard SCSI command set. SAS offers backwards-compatibility with second-generation SATA drives. SATA 3 Gbit/s drives may be connected to SAS backplanes, but SAS drives may not be connected to SATA backplanes.

SATA
Serial ATA (SATA or Serial Advanced Technology Attachment) is a computer bus interface for connecting host bus adapters to mass storage devices such as hard disk drives and optical drives. Serial ATA was designed to replace the older ATA (AT Attachment) standard (also known as EIDE). It is able to use the same low level commands, but serial ATA host-adapters and devices communicate via a high-speed serial cable over two pairs of conductors. In contrast, the parallel ATA (the redesignation for the legacy ATA specifications) used 16 data conductors each operating at a much lower speed.
SATA offers several advantages over the older parallel ATA (PATA) interface: reduced cable-bulk and cost (reduced from 80 wires to seven), faster and more efficient data transfer, and hot swapping.
The SATA host adapter is integrated into almost all modern consumer laptop computers and desktop motherboards. As of 2009, SATA has replaced parallel ATA in most shipping consumer PCs. PATA remains in industrial and embedded applications dependent on CompactFlash storage although the new CFast storage standard will be based on SATA.[2][3]

iSCSI
In computing, iSCSI (pronounced /aɪˈskʌzi/ "eye-scuzzy"), is an abbreviation of Internet Small Computer System Interface, an Internet Protocol (IP)-based storage networking standard for linking data storage facilities. By carrying SCSI commands over IP networks, iSCSI is used to facilitate data transfers over intranets and to manage storage over long distances. iSCSI can be used to transmit data over local area networks (LANs), wide area networks (WANs), or the Internet and can enable location-independent data storage and retrieval. The protocol allows clients (called initiators) to send SCSI commands (CDBs) to SCSI storage devices (targets) on remote servers. It is a popular Storage Area Network (SAN) protocol, allowing organizations to consolidate storage into data center storage arrays while providing hosts (such as database and web servers) with the illusion of locally-attached disks. Unlike traditional Fibre Channel, which requires special-purpose cabling, iSCSI can be run over long distances using existing network infrastructure.

V D I


Citrix uses the ICA (Independent Computing Architecture) protocol which is a proprietary
technology developed by the company over 10 years ago. ICA is used by both XenApp
and XenDesktop for remote desktop delivery. The company enhanced ICA with the
release of HDX (High Definition Experience) in early 2009. HDX is a set of ‘high definition’
technologies that complement the ICA protocol and allow for higher resolution
desktops/multimedia (e.g., flash video) to be delivered more efficiently and with better
quality over any network. Our industry contacts indicate that ICA is a superior protocol for
use over the WAN (i.e., for remote branches) and for connections with limited bandwidth.

VMware View primarily uses the PCoverIP protocol, which is licensed from Teradici.
PCoverIP is a remote display protocol that was developed for the purpose of delivering
virtual desktops. Mr. Viarengo of VMware opined that PCoverIP is a superior protocol
when very high resolution is required for desktops delivered over the LAN (e.g., doctors
viewing x-rays over virtual desktops). He also mentioned that while ICA has the reputation
of offering better performance over a WAN, View can also be configured to deliver similar
performance. The company commented that it is working on better communicating best
practices for how View must be configured for different network bandwidth scenarios. View
can also be configured to use Microsoft’s RDP (Remote Desktop Protocol) which appears
to be generally a less preferred technology to both ICA and PCoverIP. View continues to
use RDP to deliver VDI to certain mobile devices.

Tablet/Smart Phone Support
Citrix enables access to virtual desktops and published applications seamlessly across all
mobile devices including tablets and smart phones, through its Citrix receiver application.
Citrix receiver is a software application that sits on the mobile device and enables the user
to access their virtual desktop (XenDesktop) and published apps (XenApp).

VMware’s solution for tablet and smart phone access sometimes involves the need to use
third party applications. For example, at present, accessing View on the iPad requires the
use of Wyse Pocket Cloud client. Also, different devices use either PCoverIP or RDP, and
hence the implementation of virtual desktops across multiple devices is not consistent,
making it more difficult to support/maintain.

Storage


Storage Tiers :

0 - Special Functionality
1 - Enterprise (15,000 rpm)
2 - Modular (10,000 rpm)
3 - General Purpose (7,200 rpm SATA)
Connectivity Tier :
A - Fibre Attached
B - iSCSI Attached (not yet available)
C - NAS (not yet available)



A SAN uses the SCSI
(Small Computer Storage Interconnect) and FC (Fibre Channel) protocols to
move data over a network and store it directly to disk drives in block format

SANs use the Fibre Channel protocol

Benefits of a SAN
Removes the distance limits of SCSI-connected disks
Greater performance
Increased disk utilization
Higher availability to storage by use of multiple access paths
New disaster-recovery capabilities
Online recovery:
Reduction of servers
Increased input/output (I/O) performance and bulk data movement
Nondisruptive scalability
Storage on demand

Good for Database intensive Apps

What Makes a SAN ?
The parts: All the hardware you use to create a SAN; the switches,cables, disk arrays, and so forth
HBA , GBIC, Fiber-optic cables,
Hubs, Switches, Gateway, Router.
Storage arrays, Modular arrays, Monolithic arrays

The protocols: The languages that the parts use to talk to each other
Fibre Channel protocol, SCSI protocol

The players: The folks who build the parts

The platforms: The computer applications that benefit from using SAN

Modular arrays
Modular arrays come with shelves that hold the disk drives. Each shelf can hold between 10 to 16 drives
Modular arrays usually fit into industry-standard 19" racks
Modular arrays almostalways use two controllers with separate cache memory in each controller,and then mirror the cache between the controllers to prevent data loss. Mostmodern modular arrays have between 16 and 32GB of cache memory

Monolithic arrays
Monolithic arrays have many controllers, and those controllers can share direct access to a global memory cache (up to hundreds
of gigabytes) of fast memory. This method of sharing access to a large global or monolithic cache is why these arrays are also called monolithic.

Gigabit Interface Converter (GBIC)
The GBIC is formally known as a transceiver;it can be a transmitter and a receiver.it has a laser inside that converts billions of digital bits into light pulses to be transmitted over optical fiber.In older HBAs, the transmission device was called a Gigabit Link Module (GLM) .two kinds of GBICs, defined by the wavelength of light that the laser inside generates: short-wave (500 m) and long-wave (10 km).

Hubs use a particular version of the Fibre Channel protocol
“Protocols used in a Fibre Channel SAN,” later in this chapter) called Fibre Channel Arbitrated Loop, or FC-AL.

Standard modular(midrange-class) switches
Director-class (enterprise-class) switches
Data routers

Cables
9μm, 50μm, and 62.5μm.
When 9μm cables are used to transmit data over long distances, they’re called dark fiber cables. That’s because you cannot see the laser light being transmitted with the naked eye, and if you ever did look into one of these cables, it would fry your eyeballs. Single-mode optical signals can travel much farther than multimode signals.
Cable connectors come in two different types. An SC connector (SC stands for Subscriber connector) is the standard optical connector for 1Gbit Fibre Channel. An LC connector (LC stands for Lucent connector) is standard for 2Gbit and 4Gbit Fibre Channel cable.

N_Ports (node ports), L_Ports (loop ports), G_Ports (global ports), F_Port (fabric port), FL_Port (fabric-to-loop port), E_Port (switch-to-switch expansion port) or a T_Port ( Trunk port), NL_port (node-to-loop port),

The WWN is also sometimes called a WWPN, or World Wide Port Name.
LUN, or Logical Unit Number.
physical disk can give you only a maximum of around 150 I/O operations per second.

The disks inside a disk array are first arranged into RAID sets and then sliced up into partitions. The partitions are then assigned a LUN, and the LUN is assigned to a server in the SAN.

The WWN of the storage array is known as the World Wide Node Name or WWNN.The resulting WWN of the port on the storage array is known as the World Wide Port Name or WWPN.

no more than seven servers allocated per storage port (again, this is for each Gbps of bandwidth,but this is still a pretty good rule of thumb for even faster SAN components).Using this configuration allows those seven servers to share the connection
and therefore the bandwidth of the storage port. This is commonly called the fan-in ratio of the storage port.

Having too many servers per port also means each port has only so many I/O operations it can support at one time (the maximum queue depth of the port). Most current storage arrays support at least 256 queues per port (some support 512). So if you want each server to be able to queue up 32 I/O operations at one time (which is a good best practice), limit the number of servers
to eight per port (256/32 = 8). Most HBA vendors configure the default queue depth for their HBA drivers at 32 anyway, so this is a good default fan-in ratio for server-to-storage port.

An Infiniband adapter is called an HCA, or Host Channel Adapter; an iSCSI network card is called a TOE adapter, or TCP/IP Offload Engine adapter.

Hewlett- Packard AutoPath, SecurePath
Microsoft MPIO
Hitachi Dynamic Link Manager
EMC PowerPath
IBM RDAC, MultiPath Driver
Sun MPXIO
VERITAS Dynamic Multipathing(DMP)

✓ Dual switch
✓ Loop of switches
✓ Meshed fabric
✓ Star
✓ Core-edge

Zoning is also important because it can be used to keep storage of various servers separate from each other, keep SAN traffic localized within each zone, and separate different vendor storage arrays in the same fabric.zoning can be used as a method of making the SAN more secure.

✓ Soft zoning: Zones are identified by World Wide Name
✓ Hard zoning: Zones are identified by physical switch port