Presentation is loading. Please wait.

Presentation is loading. Please wait.

Scaleout vs. Scaleup Robert Barnes Microsoft

Similar presentations


Presentation on theme: "Scaleout vs. Scaleup Robert Barnes Microsoft"— Presentation transcript:

1 Scaleout vs. Scaleup Robert Barnes Microsoft
11/22/2018 HPTS'99 Scale Out vs Scale Up

2 HPTS'99 Scale Out vs Scale Up
Why 1BTPD? We had two main obstacles in bringing TP to Microsoft: Lack of products and infrastructure Industry mindshare and/or credibility After starting to address the first obstacle, we decided to build something to demonstrate Microsoft has a credible platform for large-scale applications CPU 50 GB Disc 5 GB RAM 11/22/2018 HPTS'99 Scale Out vs Scale Up

3 HPTS'99 Scale Out vs Scale Up
Why 1BTPD? We liked Jim Gray’s “cyber-brick” machine and wanted to build one - only the technology wasn’t there yet The closest thing to it was a collection of laptops - so we wanted to tie together 1000 laptops. We decided DebitCredit would be a good workload to run, and 10KTPS* DebitCredit was born *10 K Transactions per Second CPU 50 GB Disc 5 GB RAM 11/22/2018 HPTS'99 Scale Out vs Scale Up

4 HPTS'99 Scale Out vs Scale Up
11/22/2018 HPTS'99 Scale Out vs Scale Up

5 HPTS'99 Scale Out vs Scale Up
Why DebitCredit? We needed a DB-only benchmark Do not want to configure one million terminals A simple transaction profile to stress DTC Many little transactions Need distributed transactions to stress DTC TPC-B is dead (can not report numbers) TPC-C is not server-only (need 1 million terminals) and requires distributed data transparency So, we returned to the Datamation Benchmark (1985) 11/22/2018 HPTS'99 Scale Out vs Scale Up

6 HPTS'99 Scale Out vs Scale Up
10KTPS  1BTPD Work began on the benchmark in mid 1995 A vendor released a 30K TPM-C number Even though 10KTPS is 600Ktpm and DebitCredit is very different from TPC-C transactions, we needed a different metric tps = tpm = tph = 1 Billion Transactions Per Day* *thanks, Jim 11/22/2018 HPTS'99 Scale Out vs Scale Up

7 HPTS'99 Scale Out vs Scale Up
Scaling 11/22/2018 HPTS'99 Scale Out vs Scale Up

8 Scaleable Systems SMP and Loosely Coupled Systems
Scaleup with SMP 8P is new standard (SHV) Scaleout with Loosely Coupled Systems (LCS) using inexpensive parts SMP Super Server Departmental Server LCS of PCs or SHVs Personal System 11/22/2018 HPTS'99 Scale Out vs Scale Up

9 HPTS'99 Scale Out vs Scale Up
SMP Advantages Single system image no change to operations (relative to uniprocessor) no change to applications Simple system resources shared memory shared disk shared net Load balancing in OS kernel 8x SMP soon commodity SHV SMP Super Server Departmental Server Personal System 11/22/2018 HPTS'99 Scale Out vs Scale Up

10 HPTS'99 Scale Out vs Scale Up
SMP Problems Problems: More then 8 processors not a commodity today More than 16 processors may require partitioning, scale out needed anyway scale-down problem (starter systems expensive or fork lift upgrade) Potential single point of failure Eventually stops scaling SMP Super Server Departmental Server Personal System 11/22/2018 HPTS'99 Scale Out vs Scale Up

11 SMP Performance Diminishing returns
Linear extrapolation for 16 11/22/2018 HPTS'99 Scale Out vs Scale Up

12 HPTS'99 Scale Out vs Scale Up
LCS Advantages Advantages: No hardware limit to scale, given a scaleable interconnect and an application that partitions Uses high volume,commodity components No single point of failure Upgrade by incremental growth Balancing Load Clients and Servers made from the same stuff. Inexpensive: Built with commodity components Fault tolerance: Spare modules mask failures Modular growth grow by adding small modules Common Function 11/22/2018 HPTS'99 Scale Out vs Scale Up

13 HPTS'99 Scale Out vs Scale Up
LCS Problems Operations complexity When not designed for LCS, costs can be very expensive Change in system architecture; e.g. Exchange, SQL Server (relative to uniprocessor) Need additional system services Load Balancing LCS Membership Configuration replication Failure Retry (within LCS) Parallelism needed for data and applications explicit parallelism exposes partitions to the application implicit parallelism requires transparency Balancing Load Clients and Servers made from the same stuff. Inexpensive: Built with commodity components Fault tolerance: Spare modules mask failures Modular growth grow by adding small modules Common Function 11/22/2018 HPTS'99 Scale Out vs Scale Up

14 HPTS'99 Scale Out vs Scale Up
givememoney.com The givememoney.com business is brokering – their goal is to not directly handle any products Site is currently <10 stores, plan to be hundreds in a few years Books, music, video,computers,software,games,surplus It takes over 90 days to add a new store Over 90 servers make up current site Today, checkouts/day 10K Business plan calls for 500K by 2002 11/22/2018 HPTS'99 Scale Out vs Scale Up

15 HPTS'99 Scale Out vs Scale Up
givememoney.com FE 5 2 5 2 2 Cache Server ASP SSL ASP SSL FARM B FARM A Basket/Ad/Surplus ASP File Server SQL Product Server ASP File Server SQL Product Server Receipt/Fulfillment Games/Music Videos Games/Music Videos Monitor and cache Comp/Soft Books Comp/Soft Books Music Music Search Servers Search Servers 11/22/2018 HPTS'99 Scale Out vs Scale Up

16 HPTS'99 Scale Out vs Scale Up
givememoney.com “We don’t have a scaling problem – if we need more capacity, we just add a server. We have a management problem…” 11/22/2018 HPTS'99 Scale Out vs Scale Up

17 Scale Out Business Need
givememoney.com is at 10K orders/day, business plan is to grow to 500K by 2002 Moore’s law says scale improves by 4X in 3 years, givememoney.com wants 50X , over 10X Moore’s law. Even if you believe Moore’s law solves the growth problem, it requires many forklift upgrades We still have to scale up – as it minimizes the number of servers needed… Fantasy element….. 11/22/2018 HPTS'99 Scale Out vs Scale Up

18 HPTS'99 Scale Out vs Scale Up
Consolidation Not consolidation vs. scale out Consolidation motivations Business Site (Space) Operations Storage Reasons for never wanting to have 1 huge system Single point of failure Dis-economies of scale up Mixed workload has operations challenges similar to scale out Difficulty of dealing with growth 11/22/2018 HPTS'99 Scale Out vs Scale Up

19 Scaleout is a software challenge
Scaleup is hardware architecture & OEM driven Optimized by Software e.g. minimize locking, parallel threads, reduce contention on critical sections, etc. Ultimately, scale up begins to look like scale out (partitioning) Scaleout is software architecture driven Optimized by hardware e.g. SAN (system area nets), low latency/high bandwidth interconnects Primarily an application design and operational management problem Possible today, but mostly left to application design/development and ad hoc management tools 11/22/2018 HPTS'99 Scale Out vs Scale Up

20 HPTS'99 Scale Out vs Scale Up
Summary Scaleup begins to look like scaleout as you add processors Scaleout with commodity hardware is economically compelling Every large successful web site uses scaleout – it is practical, but hard Operations complexity is the key barrier to scaleout 11/22/2018 HPTS'99 Scale Out vs Scale Up

21 HPTS'99 Scale Out vs Scale Up
CyberWall 11/22/2018 HPTS'99 Scale Out vs Scale Up


Download ppt "Scaleout vs. Scaleup Robert Barnes Microsoft"

Similar presentations


Ads by Google