Scribd
Upload
7 views

4.3_amazon_dynamodb

Uploaded by

hungvdsoict
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
7 views

4.3_amazon_dynamodb

Uploaded by

hungvdsoict
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 16

Amazon DynamoDB

Viet-Trung Tran
Amazon DynamoDB
• Simple interface
• Key/value store
• Sacrifice strong consistency for availability
• “always writeable” data store
• no updates are rejected due to failures or concurrent writes
• Conflict resolution is executed during read instead of write
• An infrastructure within a single administrative domain where all
nodes are assumed to be trusted.

2
Design consideration
• Incremental scalability
• Symmetry
• Every node in Dynamo should have the same set of responsibilities as its peers.
• Decentralization
• In the past, centralized control has resulted in outages and the goal is to avoid
it as much as possible
• Heterogeneity
• This is essential in adding new nodes with higher capacity without having to
upgrade all hosts at once

3
System architecture
• Partitioning
• High Availability for writes
• Handling temporary failures
• Recovering from permanent failures
• Membership and failure detection

4
Partition algorithm
• Consistent hashing: the output range of a hash function is treated as a
fixed circular space or “ring”
• DynamoDB is a zero-hop DHT

Grand challenge: every nodes must maintain an up-to-date


view of the ring! How? 5
Virtual nodes
• Each node can be responsible for more than one virtual node.
• Each physical node has multiple virtual nodes
• More powerful machines have more virtual nodes
• Distribute virtual nodes across the ring
• Advantages of using virtual nodes
• If a node becomes unavailable, the load handled by this node is evenly
dispersed across the remaining available nodes.
• When a node becomes available again, or a new node is added to the system,
the newly available node accepts a roughly equivalent amount of load from
each of the other available nodes.
• The number of virtual nodes that a node is responsible can decided based on
its capacity, accounting for heterogeneity in the physical infrastructure.

6
Replication
• Each data item is replicated at N hosts.
• N is the “preference list”: The list of nodes that is responsible for storing a
particular key.

7
Quorum
• N: total number of replicas per each key/value pair
• R: minimum number of nodes that must participate in a sucessful
reading
• W: minimum number of nodes that must participate in a sucessful
writing
• Quorum-like system
• R+W>N
• In this model, the latency of a get (or put) operation is dictated by the slowest
of the R (or W) replicas. For this reason, R and W are usually configured to be
less than N, to provide better latency.

8
Temporary failures: Sloppy quorum and hinted
handoff

• Assume N = 3. When B is
temporarily down or
unreachable during a write,
send replica to E.
• E is hinted that the replica
belongs to B and it will deliver
to B when B is recovered.
• Again: “always writeable”

9
Replica synchronization
• Merkle tree
• a hash tree where leaves are hashes of the
values of individual keys
• Parent nodes higher in the tree are hashes
of their respective children
• Advantage of Merkle tree
• Each branch of the tree can be checked
independently without requiring nodes to
download the entire tree
• Help in reducing the amount of data that
needs to be transferred while checking for
inconsistencies among replicas

10
Data versioning
• A put() call may return to its caller before the update has been applied
at all the replicas
• A get() call may return many versions of the same object.
• Key Challenge: distinct version sub-histories - need to be reconciled.
• Solution: uses vector clocks in order to capture causality between different
versions of the same object.

11
Vector clock
• A vector clock is a list of (node, counter) pairs.
• Every version of every object is associated with one vector clock.
• If the counters on the first object’s clock are less-than-or-equal to all of
the nodes in the second clock, then the first is an ancestor of the
second and can be forgotten.

12
Vector clock example

When the number of (node, counter)


pairs in the vector clock reaches a
threshold (say 10), the oldest pair is
removed from the clock.

13
Technical summary

Problem Technique Advantage

Partitioning Consistent Hashing Incremental Scalability


Vector clocks with reconciliation Version size is decoupled from
High Availability for writes
during reads update rates.

Sloppy Quorum and hinted Provides high availability and


Handling temporary failures
handoff durability guarantee when some
of the replicas are not available.

Recovering from permanent Synchronizes divergent replicas


Anti-entropy using Merkle trees
failures in the background.

Preserves symmetry and avoids


having a centralized registry for
Gossip-based membership storing membership and node
Membership and failure detection
protocol and failure detection. liveness information.

14
DynamoDB sum up
• Dynamo is a highly available and scalable data store for Amazon.com’s
e-commerce platform.
• Dynamo has been successful in handling server failures, data center
failures and network partitions.
• Dynamo is incrementally scalable and allows service owners to scale
up and down based on their current request load.
• Dynamo allows service owners to customize their storage system by
allowing them to tune the parameters N, R,and W.

15
Thank you for your attention!
Q&A

You might also like