Distributed Data Analytics (WT 2017/18) - tele-TASK

แบ่งปัน
 

Fetch error

Hmmm there seems to be a problem fetching this series right now. Last successful fetch was on May 22, 2022 14:17 (5d ago)

What now? This series will be checked again in the next day. If you believe it should be working, please verify the publisher's feed link below is valid and includes actual episode links. You can contact support to request the feed be immediately fetched.

Manage series 1914092
โดย tele-TASK and Dr. Thorsten Papenbrock และถูกค้นพบโดย Player FM และชุมชนของเรา -- ลิขสิทธิ์นี้เป็นของผู้เผยแพร่ ไม่ใช่ Player FM โดยมีการสตรีมเสียงโดยตรงจากเซิร์ฟเวอร์ผู้เผยแพร่ กดปุ่มติดตามเพื่อติดตามการอัพเดทใน Player FM หรือวาง URL ฟีดนี้ไปยังแอพพอดคาสท์อื่น
The free lunch is over! Computer systems up until the turn of the century became constantly faster without any particular effort simply because the hardware they were running on increased its clock speed with every new release. This trend has changed and today's CPUs stall at around 3 GHz. The size of modern computer systems in terms of contained transistors (cores in CPUs/GPUs, CPUs/GPUs in compute nodes, compute nodes in clusters), however, still increases constantly. This caused a paradigm shift in writing software: instead of optimizing code for a single thread, applications now need to solve their given tasks in parallel in order to expect noticeable performance gains. Distributed computing, i.e., the distribution of work on (potentially) physically isolated compute nodes is the most extreme method of parallelization. Big Data Analytics is a multi-million dollar market that grows constantly! Data and the ability to control and use it is the most valuable ability of today's computer systems. Because data volumes grow so rapidly and with them the complexity of questions they should answer, data analytics, i.e., the ability of extracting any kind of information from the data becomes increasingly difficult. As data analytics systems cannot hope for their hardware getting any faster to cope with performance problems, they need to embrace new software trends that let their performance scale with the still increasing number of processing elements. In this lecture, we take a look a various technologies involved in building distributed, data-intensive systems. We discuss theoretical concepts (data models, encoding, replication, ...) as well as some of their practical implementations (Akka, MapReduce, Spark, ...). Since workload distribution is a concept which is useful for many applications, we focus in particular on data analytics.

14 ตอน