Home > Resources > Whitepapers > Virtualized Hadoop Performance with VMware vSphere® 6 on HighPerformance Servers

Virtualized Hadoop Performance with VMware vSphere® 6 on HighPerformance Servers

November 2, 2015

Large advances have been made in hardware and every level of the software stack since the virtualized Hadoop tests published in April 2013. This paper shows how to take advantage of these advances to achieve maximum performance. The cluster size remains at 32 two-processor 2U hosts; however, the processor, memory, network, and storage capabilities are all roughly doubled from those reported in the earlier paper. The performance of native and several VMware vSphere® 6 virtualized configurations were compared using the same TeraSort application suite as before. It was found that the more powerful hosts give a larger advantage to multi-VM per host configurations: virtualized TeraSort is now up to 12% faster than the optimized native configuration. The apples-to-apples case of a single virtual machine per host again shows performance close to that of native Linux. The origins of the improvements are examined and recommendations for optimal hardware and software configurations are given.

Spotlight

Advectas DACH

Advectas is specialized in corporate performance management, business intelligence, data warehousing, data mining and predictive analysis. Our customers are mainly large international companies and organizations. This provides our experts a wide knowledge of managing the complexities of multinational environments in combination with performance management. In 2015, Advectas has more than 100 employees – all passionate about Performance Management. We nurture the entrepreneurial spirit of a young company and combine it with senior experts within Business Intelligence, planning and budgeting and financial consolidation…

OTHER WHITEPAPERS

The A-Z of Master Data Management

whitePaper | January 11, 2023

The world of master data management (MDM) is a complicated place to navigate and its native speakers use a language sometimes only they themselves understand. It is packed with complex descriptions, esoteric lingo and acronyms.

Future of care: Patient-centricity with real-world predictive analytics

whitePaper | February 8, 2023

For centuries, patients have sought medical help for their ailments. Just as in the past, however, there are still many illnesses – both wellknown, widespread diseases and rare conditions – that initially cause few or inconclusive symptoms, and many patients leave the doctor’s office with an incorrect diagnosis. In addition, diseases may progress slowly or quickly depending on the individual.

Data Beyond Borders 3.0

whitePaper | July 6, 2023

Cross border data flows came to prominence under Japan’s G20 Presidency in 2019, with the Data Free Flow with Trust (DFFT) framework. Since then, the G20 Presidencies have set DFFT as a major priority in the promotion of worldwide digitisation, building the pillars that led G7 leaders to endorse and commit to a roadmap for cooperation on DFFT. Cross-border e-commerce has had a 45-fold increase1 in a decade, reaching an estimated USD2.7 trillion by 2023.2 Nearly two-thirds of global commerce is related to digital technology, with companies and governments investing an estimated USD6.8 trillion in digital transformation initiatives between 2020 and 2023.3

Analytical Data Infrastructure MarketStudy (Excerpt)

whitePaper | June 23, 2021

Analytical data infrastructure (ADI) platforms underpin key analytics models via processes like data integration, preparation, management, and storage. But with complex cloud architectures, as-a-service offerings, and innovative new analytics and process developments, how can the discerning expert choose the right ADI offerings for their business?

Cisco HyperFlex HX Data Platform

whitePaper | September 23, 2022

The Cisco HyperFlex™ HX Data Platform revolutionizes data storage for hyperconverged infrastructure deployments and makes Cisco HyperFlex Systems ready for your enterprise applications—whether they run in virtualized environments such as Microsoft Windows 2016 Hyper-V or VMware vSphere, in containerized applications using Docker and Kubernetes, or in your private or public cloud. Learn about the platform’s architecture and software-defined storage approach and how you can use it to eliminate the storage silos that complicate your data center.

Virtualized Hadoop Performance with VMware vSphere® 6 on HighPerformance Servers