This will be a collection of posts, regarding migrations in general in the first post will digging deeper in the following posts.
Published: 2013-05-09 (on www.testlabs.se/blog) Updated: 2013-05-15 Version: 1.1
Thanks for the great input and feedback: Hakim Taoussi and Magnus Göransson
Part 1: Overview
I will try to keep the first post not technical since this is more common sense then anything else. In short I want to summarize some key takeaways and recommendation to stick with, explaining them a bit more in detail below.
Planning
Information & communication
Pilot migrations
End-user training
Experience
Minimize the coexistence time
Planning
Some of you might think that… well of course we are planning. But sometimes I hear people that spend like 10-15% of their total project time for planning. I would recommend you to rethink if that’s the case, and suggest that you maybe should spend at least 50% of the time for it, maybe even more (in large projects).
What I mean with planning is to create a detailed migration plan, this should of course include estimations regarding how many users can be migrated per hour, how much data can be transferred per hour.
Basically what this means is that the planning phase should be used for planning and verifying that everything is in place and works like it’s expected to do.
For example, in the official guide from Quest Software when migrating from Domino to Exchange they calculate of 5GB/hour/migration server during good conditions.
In the real world I’ve seen throughput of 20GB/h/server. With this said, it all depends… (the consultants favorite phrase) This is one of those things that needs to be tested and verified before creating a detailed migration plan, for doing a good estimation.
Don’t forget to verify that the target environment have enough capacity, servers and storage.
Other questions that needs clear answers can be; How is users and mailboxes provisioned? During the migration, where should new mailboxes be created? Is there information in the user attributes that needs to be migrated from Domino into AD? How will the migration process work? What requirements are there?
So for the planning, think about all steps.
Information & Communication
With information I mean to inform everybody that’s involved in the project in one or another way. This would include the helpdesk and support, since these are the projects closest friends for helping and taking care of incidents.
On the other hand we have the users themselves, here I’m talking about the end-users. If the migration will impact the users in a way they are not used to, remind to inform them a couple of weeks before they are going to be migrated, with a reminding notification a couple of days when the migration will take place.
During a transition from for example, Exchange 2007 to Exchange 2010, there won’t be much impact on the users, it’s more a data transfer and updating a couple of attributes in the directory so the impact is very small. In those transition projects (it depends on the customer requirements) the needs for user reminders is not that big as the migration projects. But keep in mind, it’s better they get too much information than too little.
In large projects it’s a recommendation to place the information on public places like the restroom and the lunch room. Also inform the people on every place that’s possible, intranet, mail, letter, meeting and so on.
In short I want to say the obvious, if the information is lacking or poor, the experience from the end-user perspective will be poor. In the end this give the result of a failed project, at least from a user perspective.
Pilot migrations
From the projects I’ve been a part of I’ve learnt lots of things and gained experience. One of these things is to have a good pilot, I would recommend to divide the pilot into 3 parts.
Part 1 is the “Technical Pilot”, this would include the closest project members and/or only technical people that can handle issues and problems when they occur. Part 2 is the “Pilot 1” and this would include at least 10 users, spread throughout the organization, the more spread they are the better value would the pilot have. Part 3 is called “Pilot 2”, this is started when the “Pilot 1” phase is completed and the evaluations are done. Maybe some tweaking needs to be done before starting this stage (if there were issues and errors).
In “Pilot 2” should at least 50 people be included throughout the organization, this last Pilot phase is used for solving any issues that occurred in previous stages, this for minimizing the impact when the real migration phase will take place.
The numbers above is just examples, but might be good examples for a environment with a couple of thousand users.
Before starting with “Pilot 2” the whole migration process, how object get provisioned should be well documented. It would be a recommendation to have it documented even in the “Technical Pre-Pilot”, but my experience tells me that things are changing and somewhere during “Pilot 1” the processes are getting tested and documented.
End-user training
As this is mentioned, in some cases it might not be needed, for instance if the moved users still keeps the same Outlook client version and the impact is very low. As we all know things are changing over time with new versions and if the user used for example Outlook 2003 with Windows XP and will be upgraded to Windows 7 and Outlook 2013, there might be a reason for giving the users a training session and some documents with instructions on how things work in the new version.
If the users are migrated for example from Domino/Notes to Exchange/Outlook I would strongly recommend having training sessions were the users can attend and also bringing instructions on how things differs between Notes and Outlook, and how Outlook should be used for booking a meeting, sending a mail etc.
This for making sure that the users gets a good experience and can handle the new tools.
Minimize the coexistence time
I’m not writing this because of lack due to products out there or the functions of them.
But I’m writing this bullet for having a smoother and easier understanding, mostly for the helpdesk and the end-users. During a coexistence (freebusy/mail flow/directory synchronization) time it can be hard to troubleshoot and isolate incidents and problems. Another good reason for minimizing the coexistence time is regarding all shared resources, by minimizing the coexistence time you will reduce the impact for the end-users.
So for minimizing these hours spent on troubleshooting and the work effort everyone need to put in, I would recommend to keep the coexistence time as short as it can be, without impacting the experience or business in a bad way.
In short I would say, if things are working. Keep up a good pace for having a short coexistence time!
Experience
Last but not least, I would recommend you to select careful what project members are selected or which company that runs these kind of projects. It’s very important that they have the full understanding of what needs to be done and what impact it has for everyone involved but also the business itself.
If using Quest Software, they have a requirement of using certified people for designing, installing and configuring their products. This for making sure that the result will be good and that everyone should be satisfied with it. I’m not sure about other vendors but I think they have something similar to this model.
We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.
This will be a collection of posts, regarding migrations in general in the first post will digging deeper in the following posts.
Published: 2013-05-09 (on www.testlabs.se/blog)
Updated: 2013-05-15
Version: 1.1
Thanks for the great input and feedback: Hakim Taoussi and Magnus Göransson
Part 1: Overview
I will try to keep the first post not technical since this is more common sense then anything else.
In short I want to summarize some key takeaways and recommendation to stick with, explaining them a bit more in detail below.
Planning
Some of you might think that… well of course we are planning. But sometimes I hear people that spend like 10-15% of their total project time for planning. I would recommend you to rethink if that’s the case, and suggest that you maybe should spend at least 50% of the time for it, maybe even more (in large projects).
What I mean with planning is to create a detailed migration plan, this should of course include estimations regarding how many users can be migrated per hour, how much data can be transferred per hour.
Basically what this means is that the planning phase should be used for planning and verifying that everything is in place and works like it’s expected to do.
For example, in the official guide from Quest Software when migrating from Domino to Exchange they calculate of 5GB/hour/migration server during good conditions.
In the real world I’ve seen throughput of 20GB/h/server. With this said, it all depends… (the consultants favorite phrase) This is one of those things that needs to be tested and verified before creating a detailed migration plan, for doing a good estimation.
Don’t forget to verify that the target environment have enough capacity, servers and storage.
Other questions that needs clear answers can be;
How is users and mailboxes provisioned?
During the migration, where should new mailboxes be created?
Is there information in the user attributes that needs to be migrated from Domino into AD?
How will the migration process work?
What requirements are there?
So for the planning, think about all steps.
Information & Communication
With information I mean to inform everybody that’s involved in the project in one or another way.
This would include the helpdesk and support, since these are the projects closest friends for helping and taking care of incidents.
On the other hand we have the users themselves, here I’m talking about the end-users. If the migration will impact the users in a way they are not used to, remind to inform them a couple of weeks before they are going to be migrated, with a reminding notification a couple of days when the migration will take place.
During a transition from for example, Exchange 2007 to Exchange 2010, there won’t be much impact on the users, it’s more a data transfer and updating a couple of attributes in the directory so the impact is very small. In those transition projects (it depends on the customer requirements) the needs for user reminders is not that big as the migration projects. But keep in mind, it’s better they get too much information than too little.
In large projects it’s a recommendation to place the information on public places like the restroom and the lunch room. Also inform the people on every place that’s possible, intranet, mail, letter, meeting and so on.
In short I want to say the obvious, if the information is lacking or poor, the experience from the end-user perspective will be poor. In the end this give the result of a failed project, at least from a user perspective.
Pilot migrations
From the projects I’ve been a part of I’ve learnt lots of things and gained experience. One of these things is to have a good pilot, I would recommend to divide the pilot into 3 parts.
Part 1 is the “Technical Pilot”, this would include the closest project members and/or only technical people that can handle issues and problems when they occur.
Part 2 is the “Pilot 1” and this would include at least 10 users, spread throughout the organization, the more spread they are the better value would the pilot have.
Part 3 is called “Pilot 2”, this is started when the “Pilot 1” phase is completed and the evaluations are done. Maybe some tweaking needs to be done before starting this stage (if there were issues and errors).
In “Pilot 2” should at least 50 people be included throughout the organization, this last Pilot phase is used for solving any issues that occurred in previous stages, this for minimizing the impact when the real migration phase will take place.
The numbers above is just examples, but might be good examples for a environment with a couple of thousand users.
Before starting with “Pilot 2” the whole migration process, how object get provisioned should be well documented. It would be a recommendation to have it documented even in the “Technical Pre-Pilot”, but my experience tells me that things are changing and somewhere during “Pilot 1” the processes are getting tested and documented.
End-user training
As this is mentioned, in some cases it might not be needed, for instance if the moved users still keeps the same Outlook client version and the impact is very low. As we all know things are changing over time with new versions and if the user used for example Outlook 2003 with Windows XP and will be upgraded to Windows 7 and Outlook 2013, there might be a reason for giving the users a training session and some documents with instructions on how things work in the new version.
If the users are migrated for example from Domino/Notes to Exchange/Outlook I would strongly recommend having training sessions were the users can attend and also bringing instructions on how things differs between Notes and Outlook, and how Outlook should be used for booking a meeting, sending a mail etc.
This for making sure that the users gets a good experience and can handle the new tools.
Minimize the coexistence time
I’m not writing this because of lack due to products out there or the functions of them.
But I’m writing this bullet for having a smoother and easier understanding, mostly for the helpdesk and the end-users. During a coexistence (freebusy/mail flow/directory synchronization) time it can be hard to troubleshoot and isolate incidents and problems. Another good reason for minimizing the coexistence time is regarding all shared resources, by minimizing the coexistence time you will reduce the impact for the end-users.
So for minimizing these hours spent on troubleshooting and the work effort everyone need to put in, I would recommend to keep the coexistence time as short as it can be, without impacting the experience or business in a bad way.
In short I would say, if things are working. Keep up a good pace for having a short coexistence time!
Experience
Last but not least, I would recommend you to select careful what project members are selected or which company that runs these kind of projects. It’s very important that they have the full understanding of what needs to be done and what impact it has for everyone involved but also the business itself.
If using Quest Software, they have a requirement of using certified people for designing, installing and configuring their products. This for making sure that the result will be good and that everyone should be satisfied with it. I’m not sure about other vendors but I think they have something similar to this model.
Read more
Part 2: Prerequisites for Domino/Notes migrations
Part 3: Migrating Domino/Notes to Exchange 2013 On-premise
Part 4: Migrating User Mailboxes from Domino/Notes to Office 365
Part 5: Migrating Resources Mailboxes, Mail-In databases and Groups
Part 6: Prerequisites for Coexistence between Domino and Exchange 2013/Office 365
Part 7: Configuring Coexistence Manager for Notes with Exchange 2013 On-Prem
Part 8: Configuring Coexistence Manager for Notes with Office 365
Part 9: Prerequisites for Migration Manager
Part 10: Migrating User Mailboxes from Exchange 2003 to Exchange 2013 using Migration Manager
Part 11: Migrating User Mailboxes from Exchange On-Premise to Office 365
I hope these key takeaways gave you some good insight and some things to think about.
I would be happy to hear your comments/feedback this post.
The plan is to post a new article every second week, keep your eyes open
Regards,
Jonas
Share this post
Link to post
Share on other sites