via My Infographic Resume.

Here’s an excerpt:

A New York City subway train holds 1,200 people. This blog was viewed about 4,300 times in 2011. If it were a NYC subway train, it would take about 4 trips to carry that many people.

Click here to see the complete report.

TeamRooms Demo

However these tasks are manually trigger from the command line and do not have a password switch making an automated restart tricky.

Well here is a little gem that I hope helps someone else. Batch file are read from right to left meaning the following line would execute as:

1. Start Event Handler
2. Prompt for password
3. Echo passes password to the prompt
4. Task starts automatically

echo ‘password’  | ICMEvent.sh -d ‘database’ -u ‘user’

I believe this also works on UNIX in shell scripts.

http://publib.boulder.ibm.com/infocenter/email/v2r2m0/index.jsp?topic=%2Fcom.ibm.edc.doc%2Fedccc025.htm

I have been a very good boy this year and for Christmas I would like:

Thank you in advance, James “Your number one fan” Allen

Any one want me to pick them up one?

- Find the bottle neck, tune it, make something else the bottleneck, again, again, again…

Things to tune:

- Application layer

- P8 CE Layer

- Networking (raw network, but watch for NFS/CIFS)

- Filesystems

- Database

Most important rules:

- Exploit multi-threading

- In custom code, code carefully: Minimize server rounds trips, minimize data on the wire, find the API batching sweet spot

- Measure to estimate needed resources, monitor to see if you were right (an continue to be right over time)

Application vs CE: Should you do more processing in the application layer before ingesting the document? Consider it – you wont be able to do everything

Asynchronous vs Synchronous: synch processing has lower isolated resource cost, but asynch processing scales are better.

Auto-classification in CE ?!?!

New seamless retrieval discussed – see my previous post:

http://jamesjallen.me/2011/06/30/icc-transparent-retrieval/

Sharepoint Best Practices – new ICC deployment approach to SP front end server.

Big focus on SP – improved stubs, integration with metadata.

Delete in P8 – clear up stubs in SP now

Maintain original dates and icon for filesystems – previously the modified dates changed.

ICC 2.2 FP1 support for:

> P8 5.0 with Content Search Services

> Support for SQL server 2008 R2

> Option to disable icon changes in LN email stubbing

Roadmap:

ICC 2.2 FP2: support for p8 5.1 with CSS, IE9, Improved SP columns eg, calculated / multivalue columns, CM8 Text Indexer support for CSLD legacy itemtypes

2012 Q2 – release planned, Connectors to IBM Connections, Expand SP Support – seamless checkin, automatic stub deletion for emails & files, improved historical monitoring, improve performance for mailboxes

2013 2nd half – might be the next release

Questions:

Around Quickr and ST integration – ST there is a partner (but there is a undocument switch to write ST chats out to local filesystem as text Quickr – no plans as discussion suggested it would integrate into Connections! WOW

Single search interface – plans to merge EDM into Outlook and Notes local search.

P8 5.1 Content Search Services: IBM want to own support for indexing as in past issues occur and they rely on Verity which they cannot influence the turn around. CSS built on text indexing engine Apache Lucene (Used in dozens of IBM products). Benefits – larger index sizez, monitoring, thru’ put, scaling all better.

A new email data storage model based on xml files (sounds bit like CM8 NSE indexing )

2.2.0.2 benefits:

> Reduce storage costs

> Reduced load on FileNet P8 CE servers

> Improved indexing throughputs (1.25 – 1.5 times performance / multi lingual)

> Improved monitoring.