Google Cloud Platform Blog
Migration and Beyond - Opportunities around Windows Server 2003 EOL
Tuesday, July 14, 2015
As we’re hoping you’ve already
read
, we’ve just made Google Cloud Platform a first-class place to run your production Windows Server workloads. As we’re talking to customers concerned about Windows Server 2003 reaching
End of Life
(EOL), we’ve heard that migrations of all sorts are top of mind. We thought it would make sense to help you with resources and ideas around this transition that might make things a little easier, and a few that might make things A LOT easier.
Microsoft has a
resource that outlines a 4-step approach
to moving off of Windows Server 2003. The first two steps outlined on that page provide tools and recommendations for understanding what applications you need to move. Once you know which applications you need to move, Google Compute Engine can speed up - and reduce the cost of - the final two steps: Target and Migrate.
Google Cloud Platform can help you get off of Windows Server 2003 faster and for less money; especially if you’re having to buy gear to have enough room to move. We have some awesome insights based on talking with customers who have been using Windows Server 2008 R2 and 2012 R2 on Compute Engine during our beta that we think you’ll find useful.
Prove it works with quick, pennies-per-minute tests
Windows Server on Google Compute Engine boots fast. It’s easy to launch a Windows Server 2012 R2 instance from the
Developers Console
and be logged into the desktop in less than 7 minutes. Launching the instance takes about 5 clicks:
Windows Server on Google Compute Engine takes advantage of per-minute billing. You always pay for the first 10 minutes of an instance, and then per-minute after that.
Testing Your Application
You can use fast, pennies-per-minute access to Windows Server in Google datacenters around the world to run your application through the appropriate test suites to validate it works on the new version of Windows Server you’ve chosen.
If your application consists of multiple components, you may consider running integration tests to validate everything works together. If some of those components (Active Directory, for example) run in your data center,
VPN
allows you to securely connect your Compute Engine and on-prem networks. VPN is billed per-hour; delete it when your tests are complete and stop paying for it.
You may want to test how your application running on a new version of Windows Server performs under load. The
Distributed Load Testing
solution provides guidance and a
reference implementation
for generating and evaluating load against your application:
You can also use your own load testing framework to generate load on Compute Engine
Preemptible Instances
to reduce costs even further. In any case, when you’re done generating load you can terminate the testing infrastructure and stop paying.
User acceptance tests are also simplified with Windows Server on Google Compute Engine. You can provision a server for each tester, allowing tests to be run in parallel. You can allow RDP access to each instance while the tests are being run, and the
Chrome Remote Desktop
makes it simple for the testers to access their servers.
Moving and storing data
Whether you’re testing or actively migrating, you probably have data that your Windows Server instances on Compute Engine need to access. You might use
VPN
as described earlier to give Compute Engine instances access to an existing SAN. You could also use the
Google Cloud SDK
or
Developers Console
to copy data into
Google Cloud Storage
. Once your data is in Cloud Storage, your Compute Engine Windows Servers have high-bandwidth, low-latency access to it, and you can use an instance to create and snapshot a
persistent disk
that can be attached and used by your new VMs.
Migrating
You’ve tested and proven your application works on a supported version of Windows Server on Compute Engine. Now what? You may be tempted to scale up the new servers, migrate all the data, and go live with the update, but it’s worth considering a gradual approach with the ability to rollback in case your tests missed something. You might run the upgraded version on Compute Engine in parallel with the older version on-premises for a period of time to help identify any issues missed in testing. If you do need to roll back to the older version, simply turn off the Compute Engine infrastructure and turn it back on when you’re ready to try again.
Going all in
What if instead of only changing where your app lives, you could change
how
you run it? Some customers are using the migration away from Windows Server 2003 as a chance to embrace the cloud more completely. Rather than migrate an old data warehouse, you might consider replacing it with
BigQuery
. Just think, it’d be the last time you ever do a patch, manage a reboot, or update the operating system on your warehouse, ever! Operating a reliable queue/messaging server can be challenging, while
Google Cloud Pub/Sub
gives you access to a global, high-performance service where you only pay for what you use. If you’re using IIS to serve static content for your websites,
Google Cloud Storage
can make things dramatically easier, and substantially less expensive too.
These are just a few of the managed services offered by Google Cloud Platform that we’ve heard from customers make a big difference in TCO for their stack, but if you’re running software circa 2003, it’s probably worth taking a close look at exactly what’s available today.
All of these technologies are available to you today. There’s no commitment, you only pay for what you use, and if you are a new Google Cloud Platform customer, there’s even a
$300/60 day free trial
to help you get started. Whether you’re moving a single application from Windows Server 2003 to 2012 R2, or you’re considering a cloud-optimized architecture as part of your move, we hope that easy access to Windows Server environments on Google Cloud Platform helps make your migration easy and leaves you with no surprises.
- Posted by Miles Ward and Evan Brown, Global Head of Solutions and Solutions Architect, Google Cloud Platform
No comments :
Post a Comment
Don't Miss Next '17
Use promo code NEXT1720 to save $300 off general admission
REGISTER NOW
Free Trial
GCP Blogs
Big Data & Machine Learning
Kubernetes
GCP Japan Blog
Labels
Announcements
56
Big Data & Machine Learning
91
Compute
156
Containers & Kubernetes
36
CRE
7
Customers
90
Developer Tools & Insights
80
Events
34
Infrastructure
24
Management Tools
39
Networking
18
Open Source
105
Partners
63
Pricing
24
Security & Identity
23
Solutions
16
Stackdriver
19
Storage & Databases
111
Weekly Roundups
16
Archive
2017
Feb
Jan
2016
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2015
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2014
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2013
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2012
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2011
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2010
Dec
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2009
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Mar
Feb
Jan
2008
Dec
Nov
Oct
Sep
Aug
Jul
Jun
May
Apr
Feed
Subscribe by email
Technical questions? Check us out on
Stack Overflow
.
Subscribe to
our monthly newsletter
.
Google
on
Follow @googlecloud
Follow
Follow
No comments :
Post a Comment