Computer Science > Distributed, Parallel, and Cluster Computing
[Submitted on 29 Jan 2024 (v1), last revised 16 Feb 2024 (this version, v2)]
Title:Pilotfish: Distributed Transaction Execution for Lazy Blockchains
View PDFAbstract:Pilotfish is the first scale-out blockchain execution engine able to harness any degree of parallelizability existing in its workload. Pilotfish allows each validator to employ multiple machines, named ExecutionWorkers, under its control to scale its execution layer. Given a sufficiently parallelizable and compute-intensive load, the number of transactions that the validator can execute increases linearly with the number of ExecutionWorkers at its disposal. In addition, Pilotfish maintains the consistency of the state, even when many validators experience simultaneous machine failures. This is possible due to the meticulous co-design of our crash-recovery protocol which leverages the existing fault tolerance in the blockchain's consensus mechanism. Finally, Pilotfish can also be seen as the first distributed deterministic execution engine that provides support for dynamic reads as transactions are not required to provide a fully accurate read and write set. This loosening of requirements would normally reduce the parallelizability available by blocking write-after-write conflicts, but our novel versioned-queues scheduling algorithm circumvents this by exploiting the lazy recovery property of Pilotfish, which only persists consistent state and re-executes any optimistic steps taken before the crash. In order to prove our claims we implemented the common path of Pilotfish with support for the MoveVM and evaluated it against the parallel execution MoveVM of Sui. Our results show that Pilotfish provides good scalability up to 8 ExecutionWorkers for a variety of workloads. In computationally-heavy workloads, Pilotfish's scalability is linear.
Submission history
From: Igor Zablotchi [view email][v1] Mon, 29 Jan 2024 16:50:43 UTC (1,880 KB)
[v2] Fri, 16 Feb 2024 18:58:59 UTC (1,883 KB)
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.