Slog

Ren, Kaijun; Li, Dennis; Abadi, Daniel J.

doi:10.14778/3342263.3342647

Cited by 33 publications

(5 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Reads return the value written to a specified log position. Many applications can be implemented with shared logs, e.g., producer-consumer queues and logging [24,51].…”

Section: Making Corfu Scalable: Corfu-masonmentioning

confidence: 99%

“…However, as more shards are added, the all-to-all communication within a copy of the database will become a bottleneck, halting scalability. Eris [37], Calvin [54], vCorfu [56], Tango [4], and other distributed databases [2,42,49,51,59,60] provide a higherlevel abstraction than MASON. It is harder for services to build efficient, specialized implementations over the distributed database abstraction compared to the multi-sequence abstraction.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

[Solution] Mason: Scalable, Contiguous Sequencing for Building Consistent Services

Hodsdon¹,

Stavrinos²,

Katz-Bassett³

et al. 2023

Journal of Systems Research

View full text Add to dashboard Cite

This paper introduces a contiguous multi-sequence abstraction for building consistent, scalable services. The multi-sequence abstraction is useful when dealing with sharded datastores; the contiguity property strengthens the multi-sequence abstraction to guarantee no gaps in the sequence can be left even in the presence of failures. The outcome is an increased simplicity in developing consistent services, which is crucial for practical systems. The paper also presents MASON, a system that implements the multi-sequence abstraction and solves the scalability problem of a single monolithic sequencer. For the sake of evaluating its performance, MASON has been integrated into a distributed shared log based on CORFU, and a distributed prototype of the coordination service ZooKeeper. Overall, reviewers found the problem very relevant and timely. The presented solution is well presented, and well evaluated.

show abstract

“…Reads return the value written to a specified log position. Many applications can be implemented with shared logs, e.g., producer-consumer queues and logging [24,51].…”

Section: Making Corfu Scalable: Corfu-masonmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

[Solution] Mason: Scalable, Contiguous Sequencing for Building Consistent Services

Hodsdon¹,

Stavrinos²,

Katz-Bassett³

et al. 2023

Journal of Systems Research

View full text Add to dashboard Cite

show abstract

“…There are also works that mitigate the cost of distributed transactions but impose other constrains. For example, some mandate determinism [30,41,57,68], and are limited to noninteractive transactions that require the read/write sets of all transactions to be known prior to execution [58]. Others adopt epoch-based designs to amortize the cost of commit across several transactions [16,42,43].…”

Section: Related Workmentioning

confidence: 99%

“…Facebook's Akkio [7] splits data in -shards which migrates across datacenters to leverage locality in workloads. Similarly, SLOG [57] deploys a periodic remastering scheme over a deterministic database to reduce across-datacenter round-trips, but mandates coordination within a datacenter. Other works also exploit locality to reduce across-datacenter round-trips [25,67,74].…”

Section: Related Workmentioning

confidence: 99%

Zeus

Katsarakis

Tan

et al. 2021

Proceedings of the Sixteenth European Conference on Computer Systems

View full text Add to dashboard Cite

State-of-the-art distributed in-memory datastores (FaRM, FaSST, DrTM) provide strongly-consistent distributed transactions with high performance and availability. Transactions in those systems are fully general; they can atomically manipulate any set of objects in the store, regardless of their location. To achieve this, these systems use complex distributed transactional protocols. Meanwhile, many workloads have a high degree of locality. For such workloads, distributed transactions are an overkill as most operations only access objects located on the same server -if sharded appropriately.In this paper, we show that for these workloads, a singlenode transactional protocol combined with dynamic object re-sharding and asynchronously pipelined replication can provide the same level of generality with better performance, simpler protocols, and lower developer effort. We present Zeus, an in-memory distributed datastore that provides general transactions by acquiring all objects involved in the transaction to the same server and executing a single-node transaction on them. Zeus is fault-tolerant and stronglyconsistent. At the heart of Zeus is a reliable dynamic object sharding protocol that can move 250K objects per second per server, allowing Zeus to process millions of transactions per second and outperform more traditional distributed transactions on a wide range of workloads that exhibit locality.

show abstract

“…Still, it may be more effective than conservatively aborting Tk, and is worth considering. Other serializable certifiers, for example, SLOG [36] and Tashkent [37] also proactively decide antidependency states by certifying transactions in the order of arrival.…”

Section: Lemma 10 No Decision Made By Alg 2b Can Be Contradictedmentioning

confidence: 99%

Strict Serializable Multidatabase Certification with Out-of-Order Updates

Koutanov¹

2021

Preprint

View full text Add to dashboard Cite

Multi-phase atomic commitment protocols require long-lived resource locks on the participants and introduce blocking behaviour at the coordinator. They are also pessimistic in nature, preventing reads from executing concurrently with writes. Despite their known shortfalls, multi-phase protocols are the mainstay of transactional integration between autonomous, federated systems. This paper presents a novel atomic commitment protocol, STRIDE (Speculative Transactions in Decentralised Environments), that offers strict serializable certification of distributed transactions across autonomous, replicated sites. The protocol follows the principles of optimistic concurrency control, operating on the premise that conflicting transactions are infrequent. When they do occur, conflicting transactions are identified through antidependency testing on the certifier, which may be replicated for performance and availability. The majority of transactions can be certified entirely in memory. Unlike its multi-phase counterparts, STRIDE is nonblocking, decentralised and does not mandate the use of long-lived resource locks on the participants. It also offers a flexible isolation model for read-only transactions, which can be served directly from the participant sites without undergoing certification. Also, update transactions are Φ-serializable, making the certifier immune to the recently disclosed logical timestamp skew anomaly.

show abstract

Slog

Cited by 33 publications

References 65 publications

[Solution] Mason: Scalable, Contiguous Sequencing for Building Consistent Services

[Solution] Mason: Scalable, Contiguous Sequencing for Building Consistent Services

Zeus

Strict Serializable Multidatabase Certification with Out-of-Order Updates

Contact Info

Product

Resources

About