The performance power of software combining in persistence

Abstract

The availability of Non-Volatile Main Memory (known as NVMM) enables the design of recoverable concurrent algorithms. We study the power of software combining in achieving recoverable synchronization and designing persistent data structures. Software combining is a general synchronization approach, which attempts to simulate the ideal world when executing synchronization requests (i.e., requests that must be executed in mutual exclusion). A single thread, called the combiner, executes all active requests, while the rest of the threads are waiting for the combiner to notify them that their requests have been applied. Software combining significantly decreases the synchronization cost and outperforms many other synchronization techniques in various cases.We identify three persistence principles, crucial for performance, that an algorithm's designer has to take into consideration when designing highly-efficient recoverable synchronization protocols or data structures. We illustrate how to make the appropriate design decisions in all stages of devising recoverable combining protocols to respect these principles. Specifically, we present two recoverable software combining protocols, satisfying different progress properties, that are many times faster and have much lower persistence cost than a large collection of existing persistent techniques for achieving scalable synchronization. We build fundamental recoverable data structures, such as stacks and queues, based on these protocols that outperform by far existing recoverable implementations of such data structures. We also provide the first recoverable implementation of a concurrent heap and present experiments to show that it has good performance when the size of the heap is not very large.

When Is Recoverable Consensus Harder Than Consensus?

“…In this paper, we studied solvability, without considering efficiency. A lot of research has focused on designing efficient recoverable transactional memory systems [4, 7-9, 31, 33]) and recoverable universal constructions [11,14]. Wait-free solutions appear in [11,14,31].…”

“…A lot of research has focused on designing efficient recoverable transactional memory systems [4, 7-9, 31, 33]) and recoverable universal constructions [11,14]. Wait-free solutions appear in [11,14,31]. Some [11,14] are based on existing wait-free universal constructions [12,13] for the standard shared-memory model with halting failures.…”

“…Wait-free solutions appear in [11,14,31]. Some [11,14] are based on existing wait-free universal constructions [12,13] for the standard shared-memory model with halting failures. All except [14], satisfy weaker consistency conditions than nesting-safe recoverable linearizability.…”

“…Some [11,14] are based on existing wait-free universal constructions [12,13] for the standard shared-memory model with halting failures. All except [14], satisfy weaker consistency conditions than nesting-safe recoverable linearizability. Attiya et al [3] gave a recoverable implementation of a Compare&Swap (CAS) object.…”

“…In our setting, all shared variables are non-volatile, and we use an algorithm for RC (such as the one in Section 3.1) in place of consensus. For simplicity, we use a standard assumption (as in, e.g., [2,3,10,11,14,17,18,31]): when a process recovers from a crash, it executes a recovery function. This assumption is not restrictive; we could, alternatively, add the code of the recovery function at the beginning of the universal algorithm, thus forcing every process to execute this code before it actually starts executing a new operation.…”