Tuesday, February 1, 2011

Exchange 2010: DAG Features

Hi friends,

It’s been a while since I wrote an article, busy with the office work. Let’s move on. This article gives you what is DAG and its new feature and how it is different from the previous exchange versions.

Database Availability Group is one of the most expected new features of Exchange 2010. Microsoft has invested more time in reviewing the high availability feature of the mailbox resources.

Reason for the DAG:

1. In multisite CCR Cluster solution, the complexity in case of failures are more and especially in site resilience scenarios.
2. Features like CCR, SCC and SCR in exchange 2007 was not designed for the site resilience in the multi-site environment.
3. Hard time for engineers in handling the windows cluster dependent issues in failovers in multisite environment.
So conclusion, Simple and improved method for high availability feature is indeed a must on Exchange 2010.

Features removed from Exchange 2007:

1. No CMS or EVS concept or switches
2. No Storage group.
3. Limitation on having only Clustered Mailbox sever role without any other roles installed.
4. Exchange database is no more with the server level instead moved to Organization level.
5. No need to choose for installation of clusters or non-clustered mailbox at the start, can be done after deploying the server role (can call us incremental feature for deploying).
6. No LCR, SCC, SCR, CCR (but still there is trace of SCR and CCR patterns).

Features retained from Exchange 2007:

1. Uses Enterprise edition for DAG, since it uses limited part of Windows Failover Clustering.
2. Concept of seeding between the storage group copy with the queue length, replay time, etc. are retained in DAG as well.

New features of DAG:

1. They combined the SCR + CCR and derived a framework for high availability called DAG, which will be used for all deployment scenario - local or site or disaster cases.
2. Active Manager is the brain behind the switching/failovers. It is the replacement for the Exres.dll (Exchange Cluster resource DLL) of exchange 2007. There are two components for active manager, one called PAM – Primary Active Manager which decides on the active or passive copies and the other one called SAM – Standby Active manager which detects the failover and inform the PAM to initiate the failover.
3. Incremental Deployment i.e., forming cluster prior installing the exchange 2010 is not necessary anymore.
4. Database has been changed to the organization level from the Server level.
5. Limited dependency on Windows Failover Clustering – no more exchange application related entity are carried by the windows cluster instead you will have the limited dependency of Cluster database, heartbeat and the file share witness.
6. Co-existence with other Exchange roles since Exchange 2007 Clustered mailbox doesn’t work along with any other server roles.
7. Switch/Fail-overs much quicker than in the past.
8. Backup-less: No need to have the extensive backups/backup strategy of mailbox DB of more than 3 copies.
9. Support for DAG members in separate AD sites – member of DAG can be in different AD sites, but of Course should be of same domain within the forest.
10. Change in Log shipping: instead of SMB (Server Message Block) for shipping the log files, it uses the TCP protocol.
11. Availability of Log file Encryption and Log file compression.
12. Support for Public folder database is not supported in DAG instead it uses traditional Public folder replication mechanisms.
13. Truncation lag time value in Exchange 2007 SCR has been changed from 7 days to 14 days in Exchange 2010 DAG.

Hope the above is informative.



Anonymous said...

Admiring the persistence уou ρut into уouг websіte and in deρth informatіοn you offer.
It's awesome to come across a blog every once in a while that isn't the sаme unwantеԁ rehashed
material. Wοnderful геad! I've saved your site and I'm including your RSЅ feеds to my Googlе account.

Feеl fгee to surf to my web blog flood damage restoration
Also see my webpage: tulsa water damage restoration

Anonymous said...

Hi there, everything iѕ going fine hегe and ofcοurse everу one іs shаring dаta,
that's really fine, keep up writing.

my web site :: home water damage
my web page :: water damage clean

Anonymous said...

{I hаve|I've} been {surfing|browsing} online more than {three|3|2|4} hours today, yet I never found any interesting article like yours. {It'ѕ|It is} ρretty wοrth enough foг
me. {Іn my оpinion|Personally|In mу
view}, if all {webmаѕteгs|site oωners|websі

Reviеω my wеbsite :: water damage salt lake
{{Check out|Stop by|Visit|Review|Take a look at|Have a look at|Look at|Look into} my|My|Here is my|Also visit my|Feel free to {visit|surf} my} {web page|web site|website|webpage|site|page|blog|homepage|web-site|weblog|web blog|blog post} {:|::|;|-|...|} water damage restoration contractors