Simulation and data analysis with Austin Donnelly | July 2010.

Slides:



Advertisements
Similar presentations
Cloud Service Models and Performance Ang Li 09/13/2010.
Advertisements

Windows Azure Storage Deep Dive
VSphere vs. Hyper-V Metron Performance Showdown. Objectives Architecture Available metrics Challenges in virtual environments Test environment and methods.
= Managed for YouStandalone Servers IaaSPaaSSaaS Applications Runtimes Database Operating System Virtualization Server Storage Networking Windows.
Azure Services Platform Piotr Zierhoffer. Agenda Cloud? What is Azure? Environment Basic glossary Architecture Element description Deployment.
Windows Azure for scalable compute and storage SQL Azure for relational storage for the cloud AppFabric infrastructure to connect the cloud.
Platform as a Service (PaaS)
Google AppEngine. Google App Engine enables you to build and host web apps on the same systems that power Google applications. App Engine offers fast.
Cloud Computing Systems Lin Gu Hong Kong University of Science and Technology Sept. 21, 2011 Windows Azure—Overview.
Windows Azure Storage Services Saranya Sriram, Technology Evangelist, Microsoft, India.
WINDOWS AZURE STORAGE 11 de Mayo, 2011 Gisela Torres – Windows Azure MVP Aventia-Renacimiento Twitter:
Cross Platform Mobile Backend with Mobile Services James
Windows Azure SQL Database and Storage Name Title Organization.
Windows Azure Alex BOGDAN Academic Developer Evangelist.
Windows Azure with a dash of OSS Peter Laudati Technology Evangelist Microsoft Corporation
Abstract Load balancing in the cloud computing environment has an important impact on the performance. Good load balancing makes cloud computing more.
Using Windows Azure John Donnelly Technical Evangelist Microsoft Technology Centre Thames Valley Park
Components of Windows Azure - more detail. Windows Azure Components Windows Azure PaaS ApplicationsWindows Azure Service Model Runtimes.NET 3.5/4, ASP.NET,
Introduction To Windows Azure Cloud
Windows Azure featureISO 27001SSAE 16 SOC 1 Type 2 EU Model Clauses HIPAA BAA Web Sites Virtual Machines Cloud Services Storage (Tables,
Customers Live on Windows Azure Platform
Scott Zimmerman Solutions Architect, SOA/.NET/Azure/BizTalk.
MSDN Event. WINDOWS AZURE STORAGE Windows Azure Storage Storage in the Cloud –Scalable, durable, and available –Anywhere at anytime access –Only pay.
Larisa kocsis priya ragupathy
Austin code camp 2010 asp.net apps with azure table storage PRESENTED BY CHANDER SHEKHAR DHALL
1 NETE4631 Using Google Web Services and Using Microsoft Cloud Services Lecture Notes #7.
Windows Azure Storage – Essential Cloud Storage Services Denver Cloud Computing User Group
Bring your own machines, connectivity, software, etc. Complete control Complete responsibility Static capabilities Upfront capital costs for the infrastructure.
Jimmy Narang 1. A service in the cloud has to: Be able to handle arbitrary node failures Be available all the time Be able to scale up or down on demand.
Windows Azure Tour Benjamin Day Benjamin Day Consulting, Inc.
Mostafa Abdollahi Mazandaran University Of Science And Technology January 2011.
Windows Azure Conference 2014 Deploy your Java workloads on Windows Azure.
Overview of Cloud Computing Sven Rosvall ACCU
Windows Azure Storage Cloud Computing Soup to Nuts Mike Benkovich Microsoft Corporation btlod-72.
Windows Azure Conference 2014 Designing Applications for Scalability.
AZR308. Building distributed systems on an abstraction against commodity hardware at Internet scale, composed of multiple services. Distributed System.
T.N.C.Venkata Rangan CEO, Vishwak Solutions Your Data on Cloud.
DBI313. MetricOLTPDWLog Read/Write mixMostly reads, smaller # of rows at a time Scan intensive, large portions of data at a time, bulk loading Mostly.
ICOM 6115: Computer Systems Performance Measurement and Evaluation August 11, 2006.
Building Applications with Windows Azure Storage Brad Calder Director/Architect Microsoft Corporation.
Virtual techdays INDIA │ august 2010 Building & Migrating Web applications using Windows Azure storage Ramaprasanna Chellamuthu │ Developer Evangelist;
Visual Studio Windows Azure Portal Rest APIs / PS Cmdlets US-North Central Region FC TOR PDU Servers TOR PDU Servers TOR PDU Servers TOR PDU.
Windows Azure Virtual Machines Anton Boyko. A Continuous Offering From Private to Public Cloud.
1 Common Mistakes in Performance Evaluation (1) 1.No Goals  Goals  Techniques, Metrics, Workload 2.Biased Goals  (Ex) To show that OUR system is better.
 Brad Calder Director/Architect Microsoft Corporation ES04.
Windows Azure for scalable compute and storage SQL Azure for relational storage for the cloud AppFabric infrastructure to connect the cloud.
Azure in a Day Training: Windows Azure Module 1: Windows Azure Overview Module 2: Development Environment / Portal – DEMO: Signing up for Windows Azure.
1 Neil Kidd MTC Architect - DPE NeilKidd Neil Kidd MTC Architect - DPE NeilKidd.
INFO 344 Web Tools And Development CK Wang University of Washington Spring 2014.
Technology Drill Down: Windows Azure Platform Eric Nelson | ISV Application Architect | Microsoft UK |
(re)-Architecting cloud applications on the windows Azure platform CLAEYS Kurt Technology Solution Professional Microsoft EMEA.
Windows Azure Boot CampWindowsAzureBootCamp.com. Windows Azure Boot CampWindowsAzureBootCamp.com.
Making a Difference with Azure Storage Solutions Dudu Sinai.
Microsoft Learning Ignite | May 4 – 8, 2015 | Chicago, IL Light IT up.
BlobContainerAccount sally pictures IMG001.JPG IMG002.JPG movies MOV1.AVI.
MIX 09 11/30/2017 5:54 AM © 2009 Microsoft Corporation. All rights reserved. Microsoft, Windows, Windows Vista and other product names are or may be registered.
Platform as a Service (PaaS)
Platform as a Service (PaaS)
Design considerations for storing data in the Cloud
Network Performance and Quality of Service
Windows Azure Migrating SQL Server Workloads
Using Azure Tables In this module- Learn how to store data in tables
03 | Data Storage Bruno Terkaly | Technical Evangelist
02 | Hosting Services in Windows Azure
Windows Azure 講師: 李智樺, Ruddy Lee
Outline Virtualization Cloud Computing Microsoft Azure Platform
Windows Azure Cloud.
Saranya Sriram Developer Evangelist | Microsoft
MS AZURE By Sauras Pandey.
Austin Donnelly | July 2010.
Presentation transcript:

Simulation and data analysis with Austin Donnelly | July 2010

BIG DATA Automated observations of the world

BIG SIMULATIONS Machine-generated data

Simulations Pool fire simulation, 2040 nodes on Sandia National Lab’s Red Storm supercomputer (from SC05)

HUMAN MACHINES The unwitting cyborg

Cloud Computing Resources What for? – Statistical analysis – Simulation – Mechanical Turk / ESP Game Where from? – Departmental cluster – Project based – Windows Azure

Windows Azure

Key features: – Scalable compute – Scalable storage – Pay-as-you-go: CPU, disk, network – Higher-level API: PaaS

Cloud models Software as a Service Infrastructure as a Service Platform as a Service “SaaS”“PaaS” “IaaS” CRM ERP Collaborative Application Development Web Decision Support Streaming Caching Networking FileSecurity System Mgmt Technical

MANAGE

Declarative Services

Fabric Controller Switches Highly-available Fabric Controller Out-of-band communication – hardware control In-band communication – software control WS08 Hypervisor VM Control VM Service Roles Control Agent WS08 Node can be a VM or a physical machine Load-balancers

Hardware specs Hardware: 64-bit Windows Server 2008 Choose from four different VM sizes: S: 1x 1.6GHz, medium IO, 1.75GB / 250GB M: 2x 1.6GHz, high IO, 3.5GB / 500 GB L: 4x 1.6GHz, high IO, 7GB / 1000 GB XL: 8x 1.6GHz, high IO, 14GB / 2000 GB

STORAGE Blobs, Queues, Tables

Blobs / Example: – Account – sally – Container – music – BlobName – rock/rush/xanadu.mp3 – URL: BlobContainer Account sally pictures IMG001.JPG IMG002.JPG movies MOV1.AVI

Blobs Block Blob vs. Page Blob Snapshots Copy xDrive Geo-replication: – Dublin, Amsterdam, Chicago, Texas, Singapore, Hong Kong CDN: 18 global locations

Azure Queues QueueQueue Msg 1 Msg 2 Msg 3 Msg 4 Worker Role PutMessagePutMessage Web Role GetMessage (Timeout) RemoveMessageRemoveMessage Msg 2 Msg 1 Worker Role Msg 2 POST HTTP/ OK Transfer-Encoding: chunked Content-Type: application/xml Date: Tue, 09 Dec :04:30 GMT Server: Nephos Queue Service Version 1.0 Microsoft-HTTPAPI/ b586-0df3-4e2d-ad0c-18e3892bfca2 Mon, 22 Sep :29:20 GMT Mon, 29 Sep :29:20 GMT YzQ4Yzg1MDIGM0MDFiZDAwYzEw Tue, 23 Sep :29:20GMT PHRlc3Q+dG...dGVzdD4= HTTP/ OK Transfer-Encoding: chunked Content-Type: application/xml Date: Tue, 09 Dec :04:30 GMT Server: Nephos Queue Service Version 1.0 Microsoft-HTTPAPI/ b586-0df3-4e2d-ad0c-18e3892bfca2 Mon, 22 Sep :29:20 GMT Mon, 29 Sep :29:20 GMT YzQ4Yzg1MDIGM0MDFiZDAwYzEw Tue, 23 Sep :29:20GMT PHRlc3Q+dG...dGVzdD4= DELETE ?popreceipt=YzQ4Yzg1MDIGM0MDFiZDAwYzEw DELETE ?popreceipt=YzQ4Yzg1MDIGM0MDFiZDAwYzEw

Tables Simple entity store Entity is a set of properties – PartitionKey, RowKey, Timestamp are required (PartitionKey, RowKey) defines the key PartitionKey controls the scaling – Designed for billions of rows – PartitionKey controls locality – RowKey provides uniqueness

Partitions PartitionKey (Genre) RowKey (Title) TimestampReleaseDate Action Fast & Furious…2009 Action The Bourne Ultimatum…2007 … ……… Animation Open Season 2…2009 Animation The Ant Bully…2006 PartitionKey (Genre) RowKey (Title) TimestampReleaseDate Comedy Office Space…1999 … ……… SciFi X-Men Origins: Wolverine…2009 … ……… War Defiance…2008 PartitionKey (Genre) RowKey (Title) TimestampReleaseDate Action Fast & Furious…2009 Action The Bourne Ultimatum…2007 … ……… Animation Open Season 2…2009 Animation The Ant Bully…2006 … ……… Comedy Office Space…1999 … ……… SciFi X-Men Origins: Wolverine…2009 … ……… War Defiance…2008

Tables What tables don’t do Not relational No Referential Integrity No Joins Limited Queries No Group by No Aggregations No Transactions What tables can do CheapCheap Very Scalable FlexibleFlexible DurableDurable

Scalability targets 100TB storage per account (can ask for more) Blobs: – 200GB max block-blob size – 1TB max page-blob size Tables: – max 255 properties, totalling 1MB Queues: – 8KB messages, 1 week max age

TACTICS

HPC jobs Use worker roles – Good for parameter sweeps – Increase the invisibility time (max 2hrs) Maybe web-role as front-end

Interpreters Python, Perl etc. IronPython Remember to upload runtime dlls Think about security!

Data management Blobs for large input files: – upload may take a while, hopefully one-off – 010/04/17/windows-azure-storage-explorers.aspx 010/04/17/windows-azure-storage-explorers.aspx Dump outputs to a blob Reduce output to graphable size

Azure MODIS

Azure MODIS implementation

DATA ANALYSIS

Data curation Where did your data come from? How was it processed? Do you have the original, master data? Can you regenerate derived data? – Keep the data – Keep the code – Use a revision control system

Accuracy vs. Precision Precise Not precise AccurateNot accurate X XXX X X XXX X X X X X X X XXX X

Common mistakes in eval 1/2 No goals – Or biased goals (them vs. us) Unsystematic approach – Don’t just measure stuff at random Analysis without understanding the problem – Up to 40% of effort might be in defining problems Incorrect metrics – Right metric is not always the convenient one Wrong workload Wrong technique – Measurement, simulation, emulation, analytics? Missed parameter or factor Bad experimental design – Eg factors which interact not being varied sensibly together Wrong level of detail

Common mistakes in eval 2/2 No analysis – Measurement is not the endgame – Bad analysis – No sensitivity analysis Ignoring errors Outliers: let the wrong ones in Assume no changes in the future Ignore variability: mean is good enough Too complex model Bad presentation of results Ignore social aspects Omit assumptions and limitations

Steps for a good eval 1)State goals, define boundaries 2)Select metrics 3)List system and workload parameters 4)Select factors and their values 5)Select evaluation technique 6)Select workload 7)Design and run experiments 8)Analyse and interpret the data 9)Present results. Iterate if needed.

Books

THANKS!