23
DaMoN 2011 Paper Preview Organized by Stavros Harizopoulos and Qiong Luo Athens, Greece Jun 13, 2011

DaMoN 2011 Paper Preview

  • Upload
    lita

  • View
    32

  • Download
    3

Embed Size (px)

DESCRIPTION

DaMoN 2011 Paper Preview. Organized by Stavros Harizopoulos and Qiong Luo Athens, Greece Jun 13, 2011. Preview of Afternoon Program. 13:00-15:00 Paper Session I: FLASH DISKS, FPGAS, AND SMARTPHONES 15:00-15:30 Coffee Break 15:30-17:00 Paper Session II: MODERN CPUS AND MEMORY SYSTEMS - PowerPoint PPT Presentation

Citation preview

Page 1: DaMoN 2011 Paper Preview

DaMoN 2011 Paper Preview

Organized by Stavros Harizopoulos and Qiong Luo

Athens, GreeceJun 13, 2011

Page 2: DaMoN 2011 Paper Preview

Preview of Afternoon Program• 13:00-15:00 Paper Session I: FLASH DISKS, FPGAS,

AND SMARTPHONES

• 15:00-15:30 Coffee Break

• 15:30-17:00 Paper Session II: MODERN CPUS AND MEMORY SYSTEMS

• 17:00-17:30 Coffee Break • 17:30-18:30 Panel: WHITHER HARDWARE-

SOFTWARE CO-DESIGN?

Page 3: DaMoN 2011 Paper Preview

Paper Session I: FLASH DISKS, FPGAS, AND SMARTPHONES

• Enhancing Recovery Using an SSD Buffer Pool Extension

• Towards Highly Parallel Event Processing through Reconfigurable Hardware

• QMD: Exploiting Flash for Energy Efficient Disk Arrays

• A Case for Micro-Cellstores: Energy-Efficient Data Management on Recycled Smartphones

Page 4: DaMoN 2011 Paper Preview

IBM T.J. Watson Research Center

© 2010 IBM Corporation

Enhancing Recovery Using an SSD Bufferpool ExtensionB. Bhattacharjee, C.A. Lang, G.A.Mihaila, K. A. Ross, M. Banikazemi

All prior work including “SSD Bufferpool Extensions for Database Systems”

By M. Canim, G.A.Mihaila, B. Bhattacharjee, K. A. Ross, C.A. Lang, PVLDB 2010

Focused on exploiting

Random access capability of SSDs

Latency of SSDs

Persistence of SSDs

Server

CPUs DRAM Flash SSD SSD

SSD

HDD

HDDHDD

Storage

Bufferpool

1/2

Page 5: DaMoN 2011 Paper Preview

IBM T.J. Watson Research Center

© 2010 IBM Corporation

Contribution of this work

2/2

Prior work does not retain SSD Bufferpool contents after crash/shutdown

Leverage persistence to exploit SSD Bufferpool contents for Crash recovery of a database system Normal shutdown and start

Demonstrate Shorter recovery times Improved transaction performance after recovery With minimal overheads

Page 6: DaMoN 2011 Paper Preview
Page 7: DaMoN 2011 Paper Preview
Page 8: DaMoN 2011 Paper Preview

DaMoN 2011Athens, Greece, June 13, 2011 1/2

Page 9: DaMoN 2011 Paper Preview

DaMoN 2011Athens, Greece, June 13, 2011 2/2

Page 10: DaMoN 2011 Paper Preview

A Case for Micro-CellstoresEnergy-Efficient Data Managementon Recycled Smartphones

Stavros Harizopoulos

Spiros Papadimitriou

The views contained herein are the authors' only and do not necessarily reflect the views of Hewlett-Packard or Google

1/3

Page 11: DaMoN 2011 Paper Preview

S. Papadimitriou

A Case for Micro-CellstoresEnergy-Efficient Data Management on Recycled Smartphones

>1 billion smartphones expected to become obsolete in the next 5 years

What happens to old computers, servers, cell phones?

Can we do better?

2/3

Page 12: DaMoN 2011 Paper Preview

S. Papadimitriou

A Case for Micro-CellstoresEnergy-Efficient Data Management on Recycled Smartphones

Repurpose old smartphones

Power-profile characterization of current-generation smartphone

Initial evaluation: up to 6x more efficient (vs. other “wimpy” nodes) on scan workloads

Motivate energy-efficient, sustainable solutions

3/3

Page 13: DaMoN 2011 Paper Preview

Paper Session II:

MODERN CPUS AND MEMORY SYSTEMS • Scalable Aggregation on Multicore Processors

• How to Efficiently Snapshot Transactional Data: Hardware or Software Controlled?

• Vectorization vs. Compilation in Query Execution

Page 14: DaMoN 2011 Paper Preview

1/4 DaMoN 2011

Scalable Aggregation on

Multicore Processors

Yang Ye, Kenneth Ross, Norases VesdapuntColumbia University

DaMoN 2011

Page 15: DaMoN 2011 Paper Preview

2/4 DaMoN 2011

Utilization Challenge

• What is the best way to use the shared/partitioned resources for computations like aggregation?

• Issues:• Coordination overhead of shared data structures

• Latches and/or atomic operations• Contention

• Space overhead of replicated data structures• With n threads, each thread gets 1/nth of the shared

cache and RAM• Robustness under many input data distributions

DaMoN 2011

Page 16: DaMoN 2011 Paper Preview

3/4 DaMoN 2011

Niagara vs Nehalem

• Prior work on Sun Niagara T1 and T2 machines• Some TPC benchmark winners use the T2 (!)• Many threads: high parallelism

• Do these results generalize to other architectures such as the Nehalem processor?

• Differences in:• Clock speed• Relative cost of a miss• Degree of parallelism• Memory hierarchy & consistency model• Core sophistication (pipelines, branch prediction,

etc.)DaMoN 2011

Page 17: DaMoN 2011 Paper Preview

4/4 DaMoN 2011

Architecture Dependence

• How architecture-independent can a high-performance implementation be?

DaMoN 2011

Page 18: DaMoN 2011 Paper Preview
Page 19: DaMoN 2011 Paper Preview
Page 20: DaMoN 2011 Paper Preview

1/3

Vectorization vs. Compilationin Query Execution

Juliusz SompolskiPeter BonczMarcin Zukowski

June 13th, 2011DaMoN 2011, Athens, Greece

Page 21: DaMoN 2011 Paper Preview

2/3

Vectorization vs. Compilation:get rid of interpretation

overheadVectorization processes data in blocks to

amortize interpretation overhead over multiple tuples.

JIT query compilation generates and compiles specialized program for each query remove interpretation at all.Both get rid of interpretation overhead.

Page 22: DaMoN 2011 Paper Preview

3/3

Vectorization vs. Compilation

Once we’re rid with interpretation overhead... are they worth combining?Vectorized systems could use compilation to

move to tuple-at-a-time processing without interpretation overhead in some operations.

Existing systems using JIT compilation still choose to work tuple-at-a-time. Should they sometimes switch to vector-at-a-time model?

Case studies and examples.

Page 23: DaMoN 2011 Paper Preview

Summary

• An exciting afternoon program ahead– Seven interesting papers in two sessions

• Flash disks, FPGAs, and (recycled) smartphones• Modern (multicore) CPUs and memory systems

– Panel with experts on hardware-software co-design issues