Window-based greedy contention management for transactional memory: theory and practice

Abstract

We consider greedy contention managers for transactional memory for M × N execution windows of transactions with M threads and N transactions per thread. We present, formally analyze, and experimentally evaluate three new randomized greedy contention management algorithms for transaction windows. Assuming that each transaction has duration τ and conflicts with at most C other transactions inside the window, the first algorithm Offline-Greedy produces a schedule of length O(τ ·(C+N ·log(M N ))) with high probability. The offline algorithm depends on knowing the conflict graph which evolves while the execution of the transactions progresses. The second algorithm Online-Greedy produces a schedule of length that is only a logarithmic factor worse than Offline-Greedy, but does not require knowledge of the conflict graph. The third algorithm Adaptive-Greedy is the adaptive version of the previous algorithms which produces a schedule of length asymptotically the same as with online algorithm by adaptively guessing the value of C. All of the algorithms exhibit competitive ratio very close to O(s), where s is the number of shared resources, and at the same time, our algorithms provide new non-trivial tradeoffs for greedy transaction scheduling that parameterize window sizes and transaction conflicts within the execution window. We evaluate these window-based algorithms experimentally using the sorted link list, red-black tree, skip list, and vacation benchmarks. The evaluation results confirm This paper combines and extends preliminary results that appeared in [42,43]. their benefits in practical performance throughput and other metrics such as aborts per commit ratio and execution time overhead, along with the non-trivial provable properties of the algorithms.

Impossibility Results for Distributed Transactional Memory

“…Scheduling (or contention management) problems are widely studied in (tightlycoupled) multi-core systems and several scheduling algorithms with provable upper and lower bounds, and impossibility results are given [4,6,11,33,34,35], besides other scheduling algorithms which are evaluated only experimentally [2,39,18]. These multi-core scheduling algorithms are not suitable for transaction scheduling in distributed systems as they do not typically deal with the latency (communication cost) in accessing shared resources.…”

“…TM implementations in distributed networked systems are divided into three categories: dataflow, control-flow, and hybrid-flow based implementations. In the data-flow approach [5,16,36,35,40,41], objects are mobile and move from one node to another by traversing the network. This movement is dictated by the object requests from the transactions executing at different network nodes.…”

Transactional scheduling for read-dominated workloads

“…Following the initial publication of our paper [4], Sharma et al [26,28,27] considered the competitive ratio of schedulers that decide based on a priori knowledge about the workload (duration and data set of transactions). They propose the Clairvoyant contention manager, which is O( √ s)-competitive for balanced workloads, in which the data items written by a transaction are a constant fraction of its data set [26].…”

Adaptive Versioning in Transactional Memory Systems

“…This is typically done through a contention management strategy. There is a huge amount of work in this area giving many different strategies with and without provable properties on the guarantees they provide, e.g., [26,[60][61][62][63][64][65][66][67][68]].…”

“…Any contention management technique requires at least x − 1 transactions out of the x ≥ 2 conflicted transactions to abort. There has been an extensive study on contention management and several techniques with different performance properties are available, e.g., [26,27,[60][61][62][63][64][65][66][67][68]. We use the following strategies for resolving conflict of a transaction T with transaction T j , which are discussed in detail in [26,27,65].…”