Conclusions - Cloud Data Management - page 176

scale interactive Internet services. It relies on the Paxos protocol [ 98 ], a proven
optimal fault-tolerant consensus algorithm with no requirement for a distinguished
master, for achieving synchronous wide area replication. Megastore's replication
mechanism provides a single, consistent view of the data stored in its underlying
database replicas. Megastore replication semantics is done on entity group basis,
a priori grouping of data for fast operations, basis by synchronously replicating
the group's transaction log to a quorum of replicas. In particular, it uses a write-
ahead log replication mechanism over a group of symmetric peers where any node
can initiate reads and writes. Each log append blocks on acknowledgments from a
majority of replicas, and replicas in the minority catch up as they are able. Kraska
et al. [ 164 ] have proposed the MDCC ( M ulti- D ata C enter C onsistency) commit
protocol for providing strongly consistent guarantees at a cost that is comparable to
eventually consistent protocols. In particular, in contrast to transactional consistency
two-phase commit protocol (2PC), MDCC is designed to commit transactions
in a single round-trip across data centers in the normal operational case. It also
does not require a master node so that apply reads or updates from any node in
any data center by ensuring that every commit has been received by a quorum
of replicas. It does not also impose any database partitioning requirements. The
MDCC commit protocol can be combined with different read guarantees where
the default configuration is to guarantee read committed consistency without any
lost updates. In principle, we believe that the problem of data replication and
consistency management across different data centers in the cloud environment
has, thus far, not attracted sufficient attention from the research community, and
it represents a rich direction of future research and investigation. Nawab et al. [ 186 ]
presented Message Futures , a distributed multi-datacenter transaction management
system that provides strong consistency guarantees while maintaining low commit
latency. It achieves an average commit latency of around one Round-Trip Times
(RTT). In this approach, a transaction is committed when a commit condition
on mutual information is met. The commit condition is designed to be true, at
any point in time, for any single object in at most one datacenter. The protocol
utilizes a Replicated Log (RLog) [ 233 ] to continuously share transactions and state
information among datacenters which allows a datacenter to commit transactions
without initiating a new wide-area message exchange with other datacenters and
improves the protocol's resilience to node and communication failures.
The COPS system (Clusters of Order-Preserving Servers) [ 178 ] has been
designed to provide geo-replicated and distributed data stores that support complex
online applications, such as social networks, which must provide an always on
facility where operations always complete with low latency. In particular, it
provides causal C consistency where it executes all put and get operations in
the local datacenter in a linearizable fashion, and it then replicates data across
datacenters in a causal C consistent order in the background.COPS achieves the
causal C consistency by tracking and explicitly checking that causal dependencies
are satisfied before exposing writes in each cluster.