Impact of the convergence of three worlds: High-Performance Computing, Databases, and Analytics Stefan Manegold

Slides:



Advertisements
Similar presentations
1 Copyright © 2012 Oracle and/or its affiliates. All rights reserved. Convergence of HPC, Databases, and Analytics Tirthankar Lahiri Senior Director, Oracle.
Advertisements

Daniel Schall, Volker Höfner, Prof. Dr. Theo Härder TU Kaiserslautern.
SAN DIEGO SUPERCOMPUTER CENTER at the UNIVERSITY OF CALIFORNIA; SAN DIEGO SDSC RP Update October 21, 2010.
Lecture 2: Modern Trends 1. 2 Microprocessor Performance Only 7% improvement in memory performance every year! 50% improvement in microprocessor performance.
Shimin Chen Big Data Reading Group.  Energy efficiency of: ◦ Single-machine instance of DBMS ◦ Standard server-grade hardware components ◦ A wide spectrum.
Hybrid Hard Disk Drive Radhika Patel. Basic Terms  HDD (Hard Disk Drive): storage center for data  SSD (Solid State Drive): same thing as a hard drive,
SSD (Flash-Based) Anthony Bonomi. SSD (Solid State Drive) Commercially available for only a few years Big use in laptops Released the first 512GB last.
A “Java Fun For Everyone” Interactive Quiz
1 The Problem of Power Consumption in Servers L. Minas and B. Ellison Intel-Lab In Dr. Dobb’s Journal, May 2009 Prepared and presented by Yan Cai Fall.
Gordon: Using Flash Memory to Build Fast, Power-efficient Clusters for Data-intensive Applications A. Caulfield, L. Grupp, S. Swanson, UCSD, ASPLOS’09.

Big Data and Hadoop and DLRL Introduction to the DLRL Hadoop Cluster Sunshin Lee and Edward A. Fox DLRL, CS, Virginia Tech 21 May 2015 presentation for.
1 CS : Technology Trends Ion Stoica ( September 12, 2011.
Real Parallel Computers. Modular data centers Background Information Recent trends in the marketplace of high performance computing Strohmaier, Dongarra,
Synergy.cs.vt.edu Power and Performance Characterization of Computational Kernels on the GPU Yang Jiao, Heshan Lin, Pavan Balaji (ANL), Wu-chun Feng.
Analyzing the Energy Efficiency of a Database Server Hanskamal Patel SE 521.
Buying a Laptop. 3 Main Components The 3 main components to consider when buying a laptop or computer are Processor – The Bigger the Ghz the faster the.
Dream Machine An in depth guide to designing a gaming computer.
CERN openlab Open Day 10 June 2015 KL Yong Sergio Ruocco Data Center Technologies Division Speeding-up Large-Scale Storage with Non-Volatile Memory.
Tape is Dead Disk is Tape Flash is Disk RAM Locality is King Jim Gray Microsoft December 2006 Presented at CIDR2007 Gong Show
HPC at IISER Pune Neet Deo System Administrator
Comp-TIA Standards.  AMD- (Advanced Micro Devices) An American multinational semiconductor company that develops computer processors and related technologies.
Venkatram Ramanathan 1. Motivation Evolution of Multi-Core Machines and the challenges Background: MapReduce and FREERIDE Co-clustering on FREERIDE Experimental.
Work in Progress --- Not for Publication 1 Beyond CMOS – April 12, 2010 ERD – Emerging Research Architectures Paul Franzon,
1 Advanced Storage Technologies for High Performance Computing Sorin, Faibish EMC NAS Senior Technologist IDC HPC User Forum, April 14-16, Norfolk, VA.
Enabling Technologies for Distributed and Cloud Computing Dr. Sanjay P. Ahuja, Ph.D FIS Distinguished Professor of Computer Science School of.
Different CPUs CLICK THE SPINNING COMPUTER TO MOVE ON.
AUTHORS: STIJN POLFLIET ET. AL. BY: ALI NIKRAVESH Studying Hardware and Software Trade-Offs for a Real-Life Web 2.0 Workload.
UPPMAX and UPPNEX: Enabling high performance bioinformatics Ola Spjuth, UPPMAX
Future Server and Storage Technology Brian Minick, Infrastructure Design Leader - GE.
High Performance Computing Processors Felix Noble Mirayma V. Rodriguez Agnes Velez Electric and Computer Engineer Department August 25, 2004.
Computer Hardware Sources: Discovering Computers Information & Software technology.
S&T IT Research Support 11 March, 2011 ITCC. Fast Facts Team of 4 positions 3 positions filled Focus on technical support of researchers Not “IT” for.
Module 1: Concepts of information Technology.  Central processing unit (CPU)  Hard disk  Common input and output devices  Types of memory Main Parts.
Price Performance Metrics CS3353. CPU Price Performance Ratio Given – Average of 6 clock cycles per instruction – Clock rating for the cpu – Number of.
Computer Architecture By Chris Van Horn. CPU Basics “Brains of the Computer” Fetch Execute Cycle Instruction Branching.
1 CS : Technology Trends Ion Stoica and Ali Ghodsi ( August 31, 2015.
COMPUTER BASICS HOW TO BUILD YOUR OWN PC. CHOOSING PARTS Motherboard Processor Memory (RAM) Disk drive Graphics card Power supply Case Blu-ray/DVD drive.
Computational Research in the Battelle Center for Mathmatical medicine.
PERFORMANCE STUDY OF BIG DATA ON SMALL NODES. Ομάδα: Παναγιώτης Μιχαηλίδης Αντρέας Σόλου Instructor: Demetris Zeinalipour.
Enabling Technologies for Distributed Computing Dr. Sanjay P. Ahuja, Ph.D. Fidelity National Financial Distinguished Professor of CIS School of Computing,
Presented by NCCS Hardware Jim Rogers Director of Operations National Center for Computational Sciences.
By: Eric Moreno.  What is it?  What does it do?  What impact will it have?  When will it be available?
BY ZENIFA SHARMIN ALI (970282) ARNAB MALLICK (972127) ASHUTOSH MALI (976524) ASHUTOSH RAJ (988061) DEBANJAN KUNDU (986400) COMPUTER, LAPTOP SERVER AND.
Team Wildcats By: Patrick Kelly And Jesus Flores.
Moore’s Law Electronics 19 April Moore’s Original Data Gordon Moore Electronics 19 April 1965.
Computer Performance. Hard Drive - HDD Stores your files, programs, and information. If it gets full, you can’t save any more. Measured in bytes (KB,
1 Paolo Bianco Storage Architect Sun Microsystems An overview on Hybrid Storage Technologies.
Introduction to Data Analysis with R on HPC Texas Advanced Computing Center Feb
XNAT IT Planning Chip Schweiss June 7, Basic Requirements HTTPS proxy + Tomcat.
Multi-Core CPUs Matt Kuehn. Roadmap ► Intel vs AMD ► Early multi-core processors ► Threads vs Physical Cores ► Multithreading and Multi-core processing.
Parallel Computers Today LANL / IBM Roadrunner > 1 PFLOPS Two Nvidia 8800 GPUs > 1 TFLOPS Intel 80- core chip > 1 TFLOPS  TFLOPS = floating point.
Fundamental Digital Electronics (Fall 2013)
PC Components Microprocessor - performs all computations RAM - larger RAM memory contains more data Motherboard - holds all the above components Ports.
Anshul Gandhi 347, CS building
Hardware vs. Software Question 1 What is hardware?
Solid State Disks Testing with PROOF
40% More Performance per Server 40% Lower HW costs and maintenance
Local secondary storage (local disks)
CS : Technology Trends August 31, 2015 Ion Stoica and Ali Ghodsi (
Unit 2 Computer Systems HND in Computing and Systems Development
File Processing : Storage Media
הכרת המחשב האישי PC - Personal Computer
Some challenges in heterogeneous multi-core systems
What is the maximum capacity for DDR3 RAM?
File Processing : Storage Media
Motherboard External Hard disk USB 1 DVD Drive RAM CPU (Main Memory)
What is the maximum capacity for DDR3/DDR4 RAM?
Open Source Activity Showcase Computational Storage SNIA SwordfishTM
CS 295: Modern Systems Storage Technologies Introduction
Presentation transcript:

Impact of the convergence of three worlds: High-Performance Computing, Databases, and Analytics Stefan Manegold

2 HPC (!)

3 Database

4 Analytics (?)

5 Cluster

Our new Playground & Challenge 128 Pebbles: dual-core AMD bobcat, 8GB RAM, 10 TB HDD 128 bricks: 4-core HT i7, 16GB RAM, 2 TB HDD, 1 TB SSD 256 cores, 1 TB RAM, 1.3 PB HDD 1024 cores, 2 TB RAM, 256 TB HDD, 128 TB SSD 16 rocks: 2x 8-core HT Xeon, 256 GB RAM, 4+ GPUs, 8 TB HDD, 2 TB SSD; 48 TB NAS 512 cores, 4 TB RAM, 4+ GPUs, 128 TB HDD, 32 TB SSD; 48 TB NAS 1+ diamonds: 64+ cores, 4+ TB RAM, X GPUs, Y TB SSD, Z...

7 The Memory Wall

8 Trip to memory = 1000s of instructions!

Larger, faster, cheaper and more responsive memory sub-systems + memory-optimized DBMS =...

In the multi-core age, how do larger, faster, cheaper and more responsive memory sub-systems affect data management?

In the age of larger, faster, cheaper and more responsive memory sub-systems, how do multi- (or even many-) core systems affect data management? In the multi-core age, how do larger, faster, cheaper and more responsive memory sub-systems affect data management?

Bandwidth bottleneck: How to feed all these cores? How to exploit excess CPU cycles usefully? Exploit TurboBoost: Instructions per data dependent MPL Energy consumption: WATTs per GB of DRAM vs. Flash vs. Disk System maintenance and opertation: How to bootstrap a multi-TB (PB?) RAM system? Hot-swapable DRAM?... Shopping List Excerpt