Posts

Sharepoint notes from Matt Velloso

These are some notes from Matt Velloso's session at TechEd 09 on the 14th Sept. These are just my notes created whilst watching the session and they can be used as a start point to find out more information. Myths to debunk: You can change meta data easily. Well, yes you can, however, that doesn't mean you should. So don't. If you need to change your meta data, plan it out first and make sure that your changes don't break another part of the system. There's no code so it's risk free. Cobblers. What do you think runs Sharepoint if it's not code? Always manage risk and plan for various eventualities. It's an out of the box app, we don't need to test. Sigh.... just test it already and get over the 'testing isn't necessary' mentality. Blobcache Stop going to SQL Server all the time Use Fiddler2 (HTTP debugging proxy) to check the status of your blobcache Enable blobcache in web.config How to break your blobcache: reset IIS, hit refresh

Overview of Azure

Notes from Chris Auld's presentation at TechEd 09. Again, as with the other TechEd blogs, these are just notes that I can use to find out further information when more time is available. :) Azure: High scale application architecture Consolidate traditional IT to the cloud Move servers and apps out-house Reliable hardware in the cloud Virtualize to the cloud Manage explosive growth (scale out cloud) Scale out clouds are built around disposable hardware Reliability is built using software Scale out cloud is load balanced by default Greeness - PUE (power usage effectiveness) = Total Facility Power / IT Equipment Power. Google and Microsoft are getting around 1.10 to 1.25 PUE. Intergen server room is running at about 1.6 PUE. MS cloud offering = Windows Azure, .NET Services, SQL Azure Azure sits you above the abstraction layer (IMAGE TO GO IN HERE) Compute source (IMAGE TO GO IN HERE) Load balancer is key part of Windows Azure RoleEntryPointStart() has no return value. Always whi

Master Data Management in Kilimanjaro

Notes from the TechEd 09 presentation, presented by Rob Hawthorne. MDM was going to be in Sharepoint 2010 but is now in Kilimanjaro (SQL Server 2008 R2) MDM is NOT a transactional system MDM implements rules before data is loaded Good tool when there is difficulty coordinating multiple systems CTP and TAP in second half of 2009 (CTP3) www.microsoft.com/mdm Any application can contribute, any application can consume, process is key Master data and reference data are treated as the same thing Can coordinate multiple systems, i.e. JDE, SQLServer, DB2 etc... Performance point planning no longer exists, can do similar work in MDS but with more focus When creating new project, plan for growth but start small Uses model (cube) deployment Sharepoint integration is available MDM will hook into Dynamics CRM MDM methodology - Envision, Plan, Develop, Stabilize, Deploy Process is key; must get proceses right first before building anything

Query Optimisation

Notes from Code Camp on 13th Sept. Just a few notes that will give me some ideas to check out and use to create some more beginner level presentations. Presented by Mark Souza. Slides will be available on aucklandsql.com at some point soon, probably once TechEd 09 has finished. DMV - dm_exec_query_optimizer DBCC SHOW STATISTICS - find equivalent dmv/SQL 2005+ code AUTO_UPDATE_STATISTICS - turn off on high data turnover system using sp_autostart off Play with sample sizes, especially during development stage before going to prod. Will 10% sample size give you different statistics to a 2% sample size? sp_create_plan_guide_from_cache

Virtualization

As with my other TechEd 09 notes, this will be a list of notes that I made during the presentation. These will give a start point to find out further information. Virtualization - presented by Rob Reinauer at Code Camp on the 13th September. Find out more info on how anti-virus software works with VMs and are the AV companies planning any new methods to handle VMs? Demo of Hyper-V Live Migration in SQL Server 2008 R2 (Kilimanjaro): Copies pages to memory Copies delta changes Repeats copying delta changes down to smallest change found Freezes environment Copies everything to destination child partition Takes down source child partition Brings up destination child partition Freeze downtime is generally 2-4 seconds Connections are maintained throughout process Users will not notice the change of child partition due to maintained connections, they will most likely not even realise they have been moved In SQL11 the bulk of false failovers will be removed System Centre Virtual Machine Manag

Orinoco - complex event processing

As with my other TechEd 09 notes, this will be a list of notes that I made during the presentation. These will give a start point to find out further information. Orinoco: Search online for 'stream insights' as this is the alternative name for project Orinoco Streaming data without persistence Continuous and incremental processing of event streams from multiple sources based on declarative query and pattern specs with near zero latency Process real-time in a flow flow before it even gets to the database to speed up result processing time CEP engine - mine & design Queries look for warnings in events that come across CEP engine can sit anywhere in the process

Madison - High end scale out for data warehousing

As with my other TechEd 09 notes, this will be a list of notes that I made during the presentation. These will give a start point to find out further information. Project Madision: 100's of terabytes FastTrack reference architecture - guarantees consistent throughput for high scale data warehouses Massive scale out to 100's TB Developed by DatAllegro, MS bought them a year ago and are now incorporating their software into Kilimanjaro Large tables are hash distributed Small tables and dimension tables are replicated Create DB is now a simple script for DBAs. This script triggers a system script that will set up an optimally created database. Create table/partition is now combined in one statement