en
Andy Oram

Hadoop and Spark Performance for the Enterprise

Obavesti me kada knjiga bude dodata
Da biste čitali ovu knjigu otpremite EPUB ili FB2 datoteku na Bookmate. Kako da otpremim knjigu?
Distributed computing environments, such as Hadoop and Spark, are busy places. Multiple users are submitting various jobs with various needs in an environment where resource limits are fixed and high-priority jobs cannot be distinguished from low-priority jobs. This constant resource contention creates complex performance issues for organizations--overprovisioning of resources, late jobs, and SLAs that are not met. How can distributed processing environments evolve to ensure Quality of Service (QoS)? In this report, Andy Oram explores QoS: what it is, why it is important for organizations and professionals, and how it can be implemented in the data pipeline. By looking at how operating systems and data warehouses have developed over the years, Oram offers a glimpse of what distributed processing QoS could be. He examines two systems--Quasar and Pepperdata--that aim to improve performance in distributed processing environments by using predictive profiling and real-time assessment of resource allocation.
Ova knjiga je trenutno nedostupna
22 štampane stranice
Da li već pročitali? Kakvo je vaše mišljenje?
👍👎

Na policama za knjige

  • Алексей Хмелёв
    Big Data
    • 8
fb2epub
Prevucite i otpustite datoteke (ne više od 5 odjednom)