E-Locus - Institutional Repository of the University of Crete - TeraCache: efficient Spark caching over fast storage devices

Home TeraCache: efficient Spark caching over fast storage devices

Results - Details

[Add to Basket]

Identifier

000438707

Title

TeraCache: efficient Spark caching over fast storage devices

Alternative Title

Tera Cache: αποτελεσματική αποθήκευση ενδιάμεσων δεδομένων στο SPARK σε συσκευές γρήγορης αποθήκευσης

Author

Κολοκάσης, Ιάκωβος

Thesis advisor

Πρατικάκης, Πολύβιος
Ζακκάκ, Φοίβος

Reviewer

Μπίλας, Άγγελος
Μαγκούτης, Κώστας

Abstract

Many analytics computations are dominated by iterative processing stages, executed until a convergence condition is met. To accelerate such workloads while keeping up with the exponential growth of data and the slow scaling of DRAM capacity, Spark employs off-heap caching of intermediate results. However, off-heap caching requires serialization and deserialization (serdes) of data, that add significant overhead especially with growing datasets. This thesis proposes TeraCache, an extension of the Spark data cache that avoids the need of serdes by keeping all cached data on-heap but off-memory, using memorymapped I/O (mmio). To achieve this, TeraCache extends the original JVM heap with a managed heap that resides on a memory-mapped fast storage device and is exclusively used for cached data. Preliminary results show that the TeraCache prototype can speed up Machine Learning (ML) workloads that cache intermediate results by up to 37% compared to the state-of-the-art serdes approach.

Language

English

Issue date

2021-03-26

Collection

School/Department--School of Sciences and Engineering--Department of Computer Science--Post-graduate theses

Type of Work--Post-graduate theses

Views

552

Digital Documents
	Download document View document Views : 6