MEMORY MEANS FOR THE FUTURE OF DATABASE SYSTEMS @ANDY_PAVLO 1973 - - PowerPoint PPT Presentation

@ANDY_PAVLO WHAT NON-VOLATILE

MEMORY

MEANS FOR THE FUTURE OF

DATABASE SYSTEMS

1973

1974

1978

1986

1994

2010

2016

The Future

Non Non-Vola

• latile

tile M Memor mory

1 1

• Persistent storage with byte-addressable operations.
• Fast read/write latencies.
• No difference between random vs. sequential access.

Wh What does s NVM VM mean for D DBMSs? MSs?

1 2

• Thinking of NVM as just a faster SSD is not interesting.
• We want to use NVM as permanent storage for the

database, but this has major implications.

–Operating System Support –Cloud Provider Provisioning –Database Management System Architectures

Existing Systems NVM-Only Storage Hybrid DBMS

1 3

Existing Systems NVM-Only Storage Hybrid DBMS

Chap Chapter I r I – Exis xisting ting Systems

1 4

• Investigate how existing systems perform with NVM for

write-heavy transaction processing (OLTP) workloads.

• Evaluate two types of DBMS architectures.

–Disk-oriented (MySQL) –In-Memory (H-Store)

1 5

DISK-ORIENTED

Buffer Pool Table Heap Log Snapshots

IN-MEMORY

Table Heap Log Snapshots

In Intel l Labs abs NV NVM Em Emula ulator

• r

1 6

• Instrumented motherboard that slows down access to

the memory controller with tunable latencies.

• Special assembly to emulate upcoming Xeon

instructions for flushing cache lines.

STORE STORE

L1 Cache

L2 Cache PCOMMIT

Exp xperim rimental E al Evalu aluation tion

1 7

• Compare architectures on Intel Labs NVM emulator.
• Yahoo! Cloud Serving Benchmark:

–10 million records (~10GB) –8x database / memory –Variable skew

YCSB // //

1 8

50,000 100,000 150,000 200,000

MySQL H-Store Read-Only Workload 2x Latency Relative to DRAM

SKEW AMOUNT (HIGH→LOW) TXN/SEC

YCSB // //

1 9

10,000 20,000 30,000 40,000

TXN/SEC

MySQL H-Store 50% Reads / 50% Writes Workload 2x Latency Relative to DRAM

SKEW AMOUNT (HIGH→LOW)

8x 8x La Latency

LESSONS

2 0

Logging is a major performance bottleneck.

2 1 NVM Latency does not

have a large impact. Legacy DBMSs are not

• t

prepared to run on NVM.

3

What would Larry Ellison do?

Chap Chapter II r II – NVM VM-on

• nly S

ly Stor

• rag

age

• Evaluate storage and recovery methods for a system

that only has NVM.

• Testbed DBMS with a pluggable storage engines.
• We had to build our own NVM-aware memory allocator.

2 3

Copy-on-Write

Table Heap No Logging

Log-Structured

No Table Heap Log-only Storage

DBM DBMS Ar Architec ectures es

24

In-Place

Table Heap Log + Snapshots

Copy-on-Write

Table Heap No Logging

Log-Structured

No Table Heap Log-only Storage

In In-Place Place En Engin ine

2 5

Table Heap Log Snapshots

1 2 3

UPDATE table SET val=ABC WHERE id=123

Delta Record New Tuple New Tuple

NVM

NVM VM-Op Optim timiz ized Ar d Archit chitectur tures

2 6

• Use non-volatile pointers to only record what changed

rather than how it changed.

• Be careful about how & when things get flushed from

CPU caches to NVM.

NVM VM-Awar are In In-Place Place En Engin ine

2 7

Table Heap Log

1 2 Tuple Pointers New Tuple

Log Record TxnId Pointer UPDATE table SET val=ABC WHERE id=123

Evalu aluation tion

2 8

• Testbed system using the Intel NVM hardware emulator.
• Yahoo! Cloud Serving Benchmark

–2 million records + 1 million transactions –High-skew setting

YCSB // //

2 9

400,000 800,000 1,200,000 In-Place Copy-on-Write Log-Structured

NVM-Optimized Traditional 10% Reads / 90% Writes Workload 2x Latency Relative to DRAM ↑63

63%

↑122

22%

↑50%

0%

TXN/SEC

YCSB // //

3 0

100 200 300 400 In-Place Copy-on-Write Log-Structured

NVM-Optimized Traditional 10% Reads / 90% Writes Workload 2x Latency Relative to DRAM

NVM STORES (M)

↓40%

0%

↓25

25%

↓20%

0%

YCSB // //

3 1

NVM-Optimized Traditional Elapsed time to replay log with varying log sizes 2x Latency Relative to DRAM

RECOVERY TIME (MS)

0.01 0.1 1 10 100 1000 10^3 10^4 10^5 10^3 10^4 10^5 10^3 10^4 10^5 In-Place Copy-on-Write Log-Structured

No Recovery Needed

LESSONS

3 2

Avoid block-oriented components.

2 1 Using NVM correctly

improves throughput & reduces weadown. NVM-only systems are 15-20 years away

3

What would Nikita Kahn do?

Chap Chapter III r III – Hybrid ybrid DBMS

• Design and build a new in-memory DBMS that will be

ready for NVM when it becomes available.

• Hybrid Storage + Hybrid Workloads

–DRAM + NVM oriented architecture –Fast Transactions + Real-time Analytics

3 5

Adap daptiv tive S Stor

• rage

age

3 6

Or Orig iginal D inal Data Ada dapt pted D ed Data

SELECT AVG(B) FROM myTable WHERE C < “yyy” UPDATE myTable SET A = 123, B = 456, C = 789 WHERE D = “xxx”

A B C D

A B C D

Cold Hot

A B C D

LESSONS

3 8

Peloton

The Self-Driving Database Management System

3 9

NVM Ready Query Compilation Vectorized Execution Autonomous Apache Licensed

Anthony Tomasic Todd Mowry Prashanth Menon Michael Zhang Lin Ma Matthew Perron Dana Van Aken Yingjun Wu Ran Xian Runshen Zhu Jiexi Lin Jianhong Li Ziqi Wang

http://pelotondb.org

Joy Arulraj

@ANDY_PAVLO