Efficient algorithms for persistent transactional memory

Abstract

Durable techniques coupled with transactional semantics provide to application developers the guarantee that data is saved consistently in persistent memory (PM), even in the event of a non-corrupting failure. Persistence fences and flush instructions are known to have a significant impact on the throughput of persistent transactions.In this paper we explore different trade-offs in terms of memory usage vs. number of fences and flushes. We present two new algorithms, named Trinity and Quadra, for durable transactions on PM and implement each of them in the form of a user-level library persistent transactional memory (PTM). Quadra achieves the lower bound with respect to the number of persistence fences and executes one flush instruction per modified cache line. Trinity can be easily combined with concurrency control techniques based on fine grain locking, and we have integrated it with our TL2 adaptation, with eager locking and write-through update strategy. Moreover, the combination of Trinity and TL2 into a PTM provides good scalability for data structures and workloads with a disjoint access pattern. We used this disjoint PTM to implement a key-value (KV) store with durable linearizable transactions. When compared with previous work, our TL2 KV store provides better throughput in nearly all experiments.

TL4x

“…It is able to take advantage of both byte-and block-addressable non-volatile storage. Similar to [46], we use language interposition for load and store operations spanning persistent types during transactions.…”

“…We compare TL4x against a variety of volatile data structures that are specialized for range-queries, in addition to TrinityVRTL2 PTM [46]. We apply our PTM to a skip list, a relaxed AVL tree (RAVL), and a zip tree (sequential data structures) and compare them to a skip list, binary search tree, and a citrus tree (concurrent data structures), which have been enhanced to support range queries using techniques like versioned CAS (vCAS) [55] and bundled references [41].…”

“…Values are 100 bytes and keys are 16 bytes in size. We compare TL4x against the following key-value stores: TrinityDB [46]-based on Trin-ityVRTL2 PTM, a durably linearizable transactional memory framework which maintains a volatile replica to improve the efficiency of reads, RocksDB [37]-an in-memory key-value store that persists updates to disk or PM, either synchronously or asynchronously, and pmemkv [44]-the key-value store supplied with the persistent memory development kit (PMDK). For TL4x, we run all experiments for both PM and disk persistence, and define MIN_TXN_SYNC as 10 K transactions.…”