REPLICATION
PLACEMENT ON DISK
Introduction
• When extra storage space is available on the
striping the storage system may keep extra
copies of the stored objects to enhance the
performance of the storage system.
• If any one of the copy or the original copy is
corrupted, the corrupted copy can possibly be
recovered by comparison with its replicas.
• The replication strategy thus increases
reliability of the storage system.
Advantages
• The replication strategy can have several
advantages.
1.Replication to Increase Availability
2.Replication to Reduce Network Load
3. Replication to Reduce Start-Up Latency
4. Replication to Avoid Disk Multitasking
• Redundant array of inexpensive disks has become widely
accepted in recent years. Similar to RAID disks, the
streaming RAID was proposed to serve multimedia streams.
• The objective of streaming RAID is to increase reliability and
availability of multimedia data.
• Multimedia data are large, and each stream accesses data
of an object for a long time.
• Thus, the disk containing the accessed object will become
busy for a long period of time.
• When other streams try to access other objects residing on
the same disk, the disk becomes too busy to serve them.
• As a result, new request streams will not be served until the
disk becomes free.
• This disk outage problem limits the storage system’s ability
to serve multimedia streams without degrading their
quality.
For example, an object X is stored on three disks.
Block X is divided into many blocks.
1. X3i is the 3ith block of X and it is stored on disk 1.
2. X3i+1 is the3i+1th block of X, and it is stored on disk 2.
3. X3i+2 is the 3i+2th block of X,and it is stored on disk 3
Pi,j is found as:
For example, a data block, X3i+2, is unavailable when
the object X is being accessed.
This block can be recovered from the data blocks
X3i+1, X3i, and the parity block .
Replication to Reduce Network Load
• In the Lancaster continuous media storage server, object files are replicated
according to their distance from the originating.
• If an object’s originating site is far from the local media storage server,
the object has a higher priority of being replicated in the local server.
• If an object’s originating site is close, the object has a low priority of
being replicated.
• The Lancaster storage server provides a method to differentiate the
priority of objects being replicated in the local server.
• This method provides a mechanism to evenly spread the objects across a
number of servers over a geographical area.
• The most important advantage of the Lancaster server design is that
network load can be reduced.
Replication to Reduce Start-Up
Latency
• When multimedia streams are initiated, the storage server delivers
the first part of the object to the clients.
• The client program uses these initial data to fill up the memory
buffer.
• Before enough data are received to fill the buffers,the client program
cannot start to display the stream.
• Thus, the client program needs to wait for the delivery of the initial
part of the objects.
• This waiting time is the start-up latency. The start-up latency is a
delay time that is directly observed by the user. When multimedia
objects are being accessed over a high delay network, the start-up
latency may be significant.
• Data replication is one of the approaches to reduce the start-up
latency.
• Replicating the leaders of multimedia objects on a separate disk to
reduce start up latency in constraint allocation methods
Limitation
• A limitation of the leader replication is that
this method only reduces the start-up latency.
• After a stream is started, the leader in storage
does not make additional contributions in the
delivery of the object.
Replication to Avoid Disk Multitasking
• In magnetic hard disks, the disk heads are
connected together like a hair comb.
• All the disk heads move at the same time to the
accessed tracks and cylinders.
• Each object may be stored contiguously onto a few
cylinders.
• When one of these objects is accessed, the disk
heads only need to move one long seek action.
• The disk head is then multitasked among multiple
streams.
Replication to Maintain Balance of
Space and Load
• The access bandwidth of an object is affected by access frequency
of the object and the required bandwidth to deliver the object.
• If all the objects are accessed with the same number of megabytes
per second, the access bandwidth of the objects is linearly
proportional to their access frequency.
• Replication of objects consumes both storage space and access
bandwidth.
• When a replica of an object is stored on a storage device, the
replica may be accessed by request streams.
• The replica cannot be accessed if the storage device does not have
enough bandwidth to deliver it.
• when hot objects are stored on a low bandwidth disk
• when cold objects are stored on a high bandwidth disk
• high bandwidth disks have stored many cold objects
• low bandwidth disks have already stored many hot objects
• The objective of the bandwidth-to-space ratio (BSR)
replication is to maintain the same percentage of space and
load consumptions on all storage devices.
BSRDeviation=AllocateBSR–Actual.BSR.
