PSTs - How many users can I realisitcally migrate per week? per month?
This is long, and probably something most all of you have already gone through.
I'll cut straight to the question and then provide some background info:
Question: How many users do you realistically migrate (Ingest PSTs for) on a weekly or monthly basis?
We started our EV implementation back in June. We've somewhat successfully completed our first pilot and made a ton of usability changes as a result of that pilot.
We were creating short stubs for messages after 30 days and archiving everything, and stripping attachments everywhere.
Now after the first pilot, we're archiving items older than 1 year old. We're using the full message body as the shortcut. We're not archiving tasks, notes or contacts and we're not stripping any calendar attachments. These changes should make EV all but invisible to the majority of our 9200+ users.
Our first pilot included about 225 users.
We're set to begin a 2nd pilot in a few weeks. I've been asked to work on a plan/schedule for PST ingestion. I've talked with a few EV admins and done what research I can. I know how fast the servers can ingest data, but that's only one piece to the whole puzzle.
We are going to be ingesting all PSTs and stubbing everything with full message body. Then we'll have EV delete the PSTs as they're ingested.
I know we'll be dealing with corrupt & password protected PSTs. Don't know how many yet, so I've gotten some industry average numbers and have tried to use them to setup a schedule (I think 1-4% password protected and 8-10% corrupt is average, but I don't have the numbers in front of me).
For the longest time, we've forced users to save emails to PST files, because we had mailbox quotas in place. We still mostly do, but they've increased significantly, but we haven't advertised it.
Users of course got tired of manually moving emails, so many of them created rules to move data to PST files.
Also many users always click yes to the autoarchive prompts that outlook generates. PSTs are everything - PCs and all over the network.
With 9200 users, I've got about 27,000 PSTs. Probably about 13,000 of them are under 265K, so they're 'basically' empty, but EV will still find them and ingest them somewhere.
I've also got about 2TB of data in the 13-14K valid PSTs.
I've read the data. I can ingest about 2gigs per hour. Assuming I ingest for 8 hours per day, I can ingest 16gb per day, or 2tb in about 125 days.
But, I've got 2 EV quad core servers, so I can halve that to about 62 days, or basically 2 months.
I've been given 7 months total to do PST ingestion, and I'm trying (in vain) to explain that that's not enough time. While the data could be ingested that quickly, folks were up in arms when we archived data 30 days old using a small shortcut and only once we changed the plan to 1 year with full message body did we get approval to move forward. PSTs are going to be a huge culture shock.
I work in a hospital environment and the majority of our users are not very computer savvy. If their outlook rules no longer work, it'll be a support call. If they can't find their PST files, support call. If they can't create a new PST file, support call.
Our helpdesk doesn't have time to handle these types of calls, and our field services primarily provides hardware support. 99% of the support calls for EV will be routed to the myself or one other admin. We'll still be responsible for other support issues (7 windows admins supporting about 550 windows servers).
To make matters more fun, we're running exchange 2003. Performance is OK, but I really want to move to exchange 2010 for performance gains. That, plus trying to get users to understand 20,000 messages in the inbox or 90,000 messages in the sent items folder is bad is impossible.
The size of our exchange environment will more than double with the ingestion of 2tb of PST data. (we've got about 800GB now I believe). But the number of objects in exchange will likely triple, and I'll be dealing with performance issues as well. And the number of messages per single folder - I don't even want to think about it.
Admins I've spoken with say a good target is about 100 users per week. I've also heard that's a pretty full-time gig doing just that.
I'm having a hard time getting anyone internally to listen. Again today, I've been asked to do some testing to determine ingestion rate. Ingestion rate is the only thing anyone seems to care about.
I'd love to hear some real-world data.
I'll cut straight to the question and then provide some background info:
Question: How many users do you realistically migrate (Ingest PSTs for) on a weekly or monthly basis?
We started our EV implementation back in June. We've somewhat successfully completed our first pilot and made a ton of usability changes as a result of that pilot.
We were creating short stubs for messages after 30 days and archiving everything, and stripping attachments everywhere.
Now after the first pilot, we're archiving items older than 1 year old. We're using the full message body as the shortcut. We're not archiving tasks, notes or contacts and we're not stripping any calendar attachments. These changes should make EV all but invisible to the majority of our 9200+ users.
Our first pilot included about 225 users.
We're set to begin a 2nd pilot in a few weeks. I've been asked to work on a plan/schedule for PST ingestion. I've talked with a few EV admins and done what research I can. I know how fast the servers can ingest data, but that's only one piece to the whole puzzle.
We are going to be ingesting all PSTs and stubbing everything with full message body. Then we'll have EV delete the PSTs as they're ingested.
I know we'll be dealing with corrupt & password protected PSTs. Don't know how many yet, so I've gotten some industry average numbers and have tried to use them to setup a schedule (I think 1-4% password protected and 8-10% corrupt is average, but I don't have the numbers in front of me).
For the longest time, we've forced users to save emails to PST files, because we had mailbox quotas in place. We still mostly do, but they've increased significantly, but we haven't advertised it.
Users of course got tired of manually moving emails, so many of them created rules to move data to PST files.
Also many users always click yes to the autoarchive prompts that outlook generates. PSTs are everything - PCs and all over the network.
With 9200 users, I've got about 27,000 PSTs. Probably about 13,000 of them are under 265K, so they're 'basically' empty, but EV will still find them and ingest them somewhere.
I've also got about 2TB of data in the 13-14K valid PSTs.
I've read the data. I can ingest about 2gigs per hour. Assuming I ingest for 8 hours per day, I can ingest 16gb per day, or 2tb in about 125 days.
But, I've got 2 EV quad core servers, so I can halve that to about 62 days, or basically 2 months.
I've been given 7 months total to do PST ingestion, and I'm trying (in vain) to explain that that's not enough time. While the data could be ingested that quickly, folks were up in arms when we archived data 30 days old using a small shortcut and only once we changed the plan to 1 year with full message body did we get approval to move forward. PSTs are going to be a huge culture shock.
I work in a hospital environment and the majority of our users are not very computer savvy. If their outlook rules no longer work, it'll be a support call. If they can't find their PST files, support call. If they can't create a new PST file, support call.
Our helpdesk doesn't have time to handle these types of calls, and our field services primarily provides hardware support. 99% of the support calls for EV will be routed to the myself or one other admin. We'll still be responsible for other support issues (7 windows admins supporting about 550 windows servers).
To make matters more fun, we're running exchange 2003. Performance is OK, but I really want to move to exchange 2010 for performance gains. That, plus trying to get users to understand 20,000 messages in the inbox or 90,000 messages in the sent items folder is bad is impossible.
The size of our exchange environment will more than double with the ingestion of 2tb of PST data. (we've got about 800GB now I believe). But the number of objects in exchange will likely triple, and I'll be dealing with performance issues as well. And the number of messages per single folder - I don't even want to think about it.
Admins I've spoken with say a good target is about 100 users per week. I've also heard that's a pretty full-time gig doing just that.
I'm having a hard time getting anyone internally to listen. Again today, I've been asked to do some testing to determine ingestion rate. Ingestion rate is the only thing anyone seems to care about.
I'd love to hear some real-world data.
