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/

Global: Zendy Tools annual

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

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Datacenter capacity is growing exponentially to satisfy the increasing demandfor emerging computationally-intensive applications, such as deep learning.This trend has led to concerns over datacenters' increasing energy consumptionand carbon footprint. The basic prerequisite for optimizing a datacenter'senergy- and carbon-efficiency is accurately monitoring and attributing energyconsumption to specific users and applications. Since datacenter servers tendto be multi-tenant, i.e., they host many applications, server- and rack-levelpower monitoring alone does not provide insight into their residentapplications' energy usage and carbon emissions. At the same time, currentapplication-level energy monitoring and attribution techniques are intrusive:they require privileged access to servers and require coordinated support inhardware and software, which is not always possible in cloud. To address theproblem, we design WattScope, a system for non-intrusively estimating the powerconsumption of individual applications using external measurements of aserver's aggregate power usage without requiring direct access to the server'soperating system or applications. Our key insight is that, based on an analysisof production traces, the power characteristics of datacenter workloads, e.g.,low variability, low magnitude, and high periodicity, are highly amenable todisaggregation of a server's total power consumption into application-specificvalues. WattScope adapts and extends a machine learning-based technique fordisaggregating building power and applies it to server- and rack-level powermeter measurements in data centers. We evaluate WattScope's accuracy on aproduction workload and show that it yields high accuracy, e.g., often <10%normalized mean absolute error, and is thus a potentially useful tool fordatacenters in externally monitoring application-level power usage.

WattScope: Non-intrusive Application-level Power Disaggregation in Datacenters