Browsing by Author "Rajappan, Gowtham"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Scalable Transaction Management for Partially Replicated Data in Cloud Computing Environments
    (IEEE, 2016) Tripathi, Anand; Rajappan, Gowtham
    We present here a scalable protocol for transaction management in key-value based multi-version data storage systems supporting partial replication of data in cloud and cluster computing environments. We consider here systems in which the database is sharded into partitions, a partition is replicated only at a subset of the nodes in the system, and no node contains all partitions. The protocol presented here is based on the Partitioned Causal Snapshot Isolation (PCSI) model and it enhances the scalability of that model. The PCSI protocol is scalable for update transactions which involve updating of only local partitions. However, it faces scalability limitations when transactions update non-local partitions. This limitation stems from the scheme used for obtaining update timestamps for remote partitions, causing vector clocks to grow with the system configuration size. We present here a new protocol based on the notion of sequence number escrow and address the underlying technical problems. Our experimental evaluations show that this protocol scales out almost linearly when workloads involve transactions with remote partition updates. We present here the performance of this protocol for three different workloads with varying mix of transaction characteristics.
  • Thumbnail Image
    Item
    Scalable Transactions in Partially Replicated Data Systems with Causal Snapshot Isolation
    (2015-07-14) Tripathi, Anand; Rajappan, Gowtham; Padhye, Vinit
    We present here a transaction management protocol, which enhances the Partitioned Causal Snapshot Isolation (PCSI) pro- tocol, to support scalable transactions with non-local partition writes in a partially replicated multi-version database. The PCSI protocol is scalable for update transactions that involve local read and writes. However, it faces scalability limitations with non-local partition writes, when partitions are updated from any of the sites. In PCSI, to support non-local partition writes, a ghost replica of that partition is created at the transaction execution site. It is used primarily for assigning update sequence numbers to the local transactions and does not store any data. We present here an alternate approach for supporting non-local partition writes. The update sequence number for a non-local partition update is obtained from one of the remote sites that stores a replica of that partition. This approach raises several problems in regard to stalling of transactions at the remote replica’s site. We develop here a protocol based on the notion of sequence number escrow to address these problems. Through experimental evaluations, we show that this approach for supporting non-local partition updates scales almost linearly.