Download - Store Everything Online In A Database
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 1
Store Everything
OnlineIn A Database
Jim GrayMicrosoft Research
[email protected]://research.microsoft.com/~gray/talks
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 2
Outline
•Store Everything•Online (Disk not Tape)
• In a Database
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 3
How Much is Everything?• Soon everything can be
recorded and indexed• Most bytes will never be
seen by humans.• Data summarization, trend
detection anomaly detection are key technologies
See Mike Lesk: How much information is there: http://www.lesk.com/mlesk/ksg97/ksg.html
See Lyman & Varian:
How much informationhttp://www.sims.berkeley.edu/research/projects/how-much-info/
Yotta
Zetta
Exa
Peta
Tera
Giga
Mega
KiloA BookA Book
.Movie
All LoC books(words)
All Books MultiMedia
Everything!
Recorded
A PhotoA Photo
24 Yecto, 21 zepto, 18 atto, 15 femto, 12 pico, 9 nano, 6 micro, 3 milli
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 4
1E+3
1E+4
1E+5
1E+6
1E+7
1988 1991 1994 1997 2000
disk TB growth: 112%/y
Moore's Law: 58.7%/y
ExaByte
Disk TB Shipped per Year1998 Disk Trend (J im Porter)
http://www.disktrend.com/pdf/portrpkg.pdf.
Storage capacity beating Moore’s law
3 k$/TB today (raw disk)
1k$/TB by end of 2002
Moores law 58.70% /year
Revenue 7.47%TB growth 112.30% (since 1993)
Price decline 50.70% (since 1993)
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 5
Outline
•Store Everything•Online (Disk not Tape)
• In a Database
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 6
Online Data
• Can build 1PB of NAS disk for 5M$ today
• Can SCAN (read or write) entire PB in 3 hours.• Operate it as a data pump: continuous sequential scan
• Can deliver 1PB for 1M$ over Internet– Access charge is 300$/Mbps bulk rate
• Need to Geoplex data (store it in two places).
• Need to filter/process data near the source,– To minimize network costs.
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 7
The “Absurd” Disk
• 2.5 hr scan time (poor sequential access)
• 1 access per second / 5 GB (VERY cold data)
• It’s a tape!
1 TB100 MB/s
200 Kaps
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 8
Disk vs Tape
Disk– 80 GB– 35 MBps– 5 ms seek time– 3 ms rotate latency– 3$/GB for drive
2$/GB for ctlrs/cabinet– 15 TB/rack
– 1 hour scan
Tape– 40 GB– 10 MBps– 10 sec pick time– 30-120 second seek time– 2$/GB for media
8$/GB for drive+library– 10 TB/rack
– 1 week scan
The price advantage of disk is growing the performance advantage of disk is huge!At 10K$/TB, disk is competitive with nearline tape.
GuestimatesCern: 200 TB3480 tapes2 col = 50GBRack = 1 TB=12 drives
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 90
100
200
300
400
500
Premium SAN Dell/3ware DIY
Building a Petabyte Disk Store• Cadillac ~ 500k$/TB = 500M$/PB
plus FC switches plus… 800M$/PB• TPC-C SANs (Brand PC 18GB/…) 60 M$/PB• Brand PC local SCSI 20M$/PB• Do it yourself ATA 5M$/PB
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 10
Cheap Storage and/or Balanced System
• Low cost storage (2 x 3k$ servers) 5K$ TB2x ( 800 Mhz, 256Mb + 8x80GB disks + 100MbE)
raid5 costs 6K$/TB
• Balanced server (5k$/.64 TB)– 2x800Mhz (2k$)– 512 MB – 8 x 80 GB drives (2K$)– Gbps Ethernet + switch (300$/port)– 9k$/TB 18K$/mirrored TB
2x800 Mhz512 MB
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 11
Next step in the Evolution• Disks become supercomputers
– Controller will have 1bips, 1 GB ram, 1 GBps net– And a disk arm.
• Disks will run full-blown app/web/db/os stack
• Distributed computing
• Processors migrate to transducers.
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 12
It’s Hard to Archive a PetabyteIt takes a LONG time to restore it.
• At 1GBps it takes 12 days!• Store it in two (or more) places online (on disk?).
A geo-plex• Scrub it continuously (look for errors)• On failure,
– use other copy until failure repaired, – refresh lost copy from safe copy.
• Can organize the two copies differently (e.g.: one by time, one by space)
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 13
Outline
•Store Everything•Online (Disk not Tape)
• In a Database
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 14
Why Not file = object + GREP?• It works if you have thousands of objects
(and you know them all)
• But hard to search millions/billions/trillions with GREP
• Hard to put all attributes in file name.– Minimal metadata
• Hard to do chunking right.
• Hard to pivot on space/time/version/attributes.
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 15
The Reality: it’s build vs buy
• If you use a file system you will eventually build a database system:– metadata, – Query, – parallel ops, – security,….– reorganize, – recovery, – distributed, – replication,
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 16
OK: so I’ll put lots of objects in a fileDo It Yourself Database
• Good news: – Your implementation will be
10x faster than the general purpose one easier to understand and use than the general purpose on.
• Bad news: – It will cost 10x more to build and maintain– Someday you will get bored maintaining/evolving it– It will lack some killer features:
• Parallel search• Self-describing via metadata• SQL, XML, … • Replication• Online update – reorganization• Chunking is problematic (what granularity, how to aggregate)
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 17
Top 10 reasons to put Everything in a DB1. Someone else writes the million lines of code2. Captures data and Metadata,3. Standard interfaces give tools and quick learning4. Allows Schema Evolution without breaking old apps5. Index and Pivot on multiple attributes
space-time-attribute-version….6. Parallel terabyte searches in seconds or minutes7. Moves processing & search close to the disk arm
(moves fewer bytes (qestons return datons). 8. Chunking is easier (can aggregate chunks at server).9. Automatic geo-replication 10. Online update and reorganization. 11. Security 12. If you pick the right vendor, ten years from now, there will
be software that can read the data.
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 18
DB Centric Examples
• TerraServer– All images and all data in the database (chunked as small
tiles).www.TerraServer.Microsoft.com/
– http://research.microsoft.com/~gray/Papers/MSR_TR_99_29_TerraServer.doc
• SkyServer & Virtual Sky– Both image and semantic data in a relational store.– Parallel search & NonProcedural access are important.– http://research.microsoft.com/~gray/Papers/MS_TR_99_30_Sloan_Digital_Sky_Survey.doc
– http://dart.pha.jhu.edu/sdss/getMosaic.asp?Z=1&A=1&T=4&H=1&S=10&M=30– http://virtualsky.org/servlet/Page?F=3&RA=16h+10m+1.0s&DE=
%2B0d+42m+45s&T=4&P=12&S=10&X=5096&Y=4121&W=4&Z=-1&tile.2.1.x=55&tile.2.1.y=20
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 19
OK… Why don’t they use our stuff?
• Wrong metaphor: HDF with hyper-slab is better match.
• Impedence match: getting stuff in/out of DB is too hard
• We sold them OODBs and they did not work (unreliable, poor performance, no tools).
• …
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 20
So, why will the future be different?
• They have MUCH more data (10^8 files?)
• Java / C# eases impedance mismatch: rowsets == ragged arrays.
• Tools are better– Optimizers are better– CPU and disk parallelism actually works now– Statistical packages are better.
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 21
Outline
•Store Everything•Online (Disk not Tape)
• In a Database
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 22
But… The title of the talk was…“The Future of
Distributed Database Systems”
Nobody wants to share his database.
blocks, files, tables are wrong abstraction for networks.(too low level)
“Objects are the right abstraction”
So, UDDI / WSDL / SOAP is the solution (not SQL)
XML is the wire format, XLANG is the workflow protocol, Query will be in there somewhere.
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 23
DDB technology GREAT in a Cluster
• Uniform architecture
• Trust among nodes
• High bandwidth-low latency communication
• Programs have single system image
• Queries run in parallel
• Global optimizer does query decomposition
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 24
But in a Distributed System
• Heterogenous architecture makes query planning much harder
• No trust
• Communication is slow and expensive (minimize it).
Higher level abstraction to minimize round trips
http://research.microsoft.com/~gray/talks/Gray_GriPhyN.ppt 25
DDB the Trust Issue• Customers serve
themselves• Follow the rules posted
on the door• No Overhead, no staff!
• Clerks serve Customers • Take order, fill order, fill out
invoice, collect money. • Overhead: staff, training, rules,
…
DDB Grocery
• Customers serve themselves
• Follow the rules posted on the dorr
Client/Server Groceries