English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

While (1) serverless computing is emerging as a popular form of cloudexecution, datacenters are going through major changes: (2) storagedissaggregation in the system infrastructure level and (3) integration ofdomain-specific accelerators in the hardware level. Each of these three trendsindividually provide significant benefits; however, when combined the benefitsdiminish. Specifically, the paper makes the key observation that for serverlessfunctions, the overhead of accessing dissaggregated persistent storageovershadows the gains from accelerators. Therefore, to benefit from all thesetrends in conjunction, we propose Domain-Specific Computational Storage forServerless (DSCS-Serverless). This idea contributes a serverless model thatleverages a programmable accelerator within computational storage to conjugatethe benefits of acceleration and storage disaggregation simultaneously. Ourresults with eight applications shows that integrating a comparatively smallaccelerator within the storage (DSCS-Serverless) that fits within its powerconstrains (15 Watts), significantly outperforms a traditional disaggregatedsystem that utilizes the NVIDIA RTX 2080 Ti GPU (250 Watts). Further, the workhighlights that disaggregation, serverless model, and the limited power budgetfor computation in storage require a different design than the conventionalpractices of integrating microprocessors and FPGAs. This insight is in contrastwith current practices of designing computational storage that are yet toaddress the challenges associated with the shifts in datacenters. In comparisonwith two such conventional designs that either use quad-core ARM A57 or aXilinx FPGA, DSCS-Serverless provides 3.7x and 1.7x end-to-end applicationspeedup, 4.3x and 1.9x energy reduction, and 3.2x and 2.3x higher costefficiency, respectively.

In-Storage Domain-Specific Acceleration for Serverless Computing