A friend of mine turned me onto this company this week and I finally had some time to poke around the site. I had to smile when I saw what the solution does, because anyone who has ever worked in support (internal or customer facing) will understand the value.
There was a comment that was posted back on my musings about Vkernel which was 'They (Vkernel) don't do anything that a good Sys Admin can't do'. To which I say, then why are providing value to CIO's for a lot less money than a 'good Sys Admin'?
Same goes for AutoVirt. Their focus is on data migration and making sure that as data is migrated - the stuff that usually gets broken - re-naming, repointing, remebering what you did in what order, all the stuff that a 'good Sys Admin' can absolutely do - that the migrated data and the breadcrumb trail and detour signs stay intact, so your users don't inflict more pain on your 'good Sys Admins'.
BTW, I was *not* a 'good Sys Admin' because I was lazy and not detail oriented enough to write everything I did down someplace, nor did I have the memory that would help me remember what I did in the right order. That's why I see the value in the AutoVirt's and Vkernel's of the world - they do the stuff that I am too lazy to do and will help me keep my job in spite of myself.
Showing posts with label Server Virtualization. Show all posts
Showing posts with label Server Virtualization. Show all posts
Friday, November 21, 2008
Friday, July 25, 2008
Three Tier Virtualization
In my travels the past couple of weeks I have some interesting discussions with folks from the purely hypothetical - Where do you think Virtualization will be in a year or two - to basic blocking and tackling discussions - How do I figure out chargebacks and what should I virtualize first?
What struck me was that these are the same types of discussions that have happened in IT since I have been dealing with it and they center arount architecture, and more specifically eating the Elephant in Bites by using tiers in our approach.
So on the VDI or Virtual Desktop front I have had discussions with some fairly large companies about VDI plans. The bulk of them are going to Virtualize Windows and the two approaches seem to be to use Citrix/Xen or one of the new startup management tools to do this. Why Microsoft never provided a way to manage desktops better in the first place is beyond me - they figured out how to deliver a security patch every Tuesday a while ago and RDS (Remote Desktop Support) has been around a while too. Alas - I am off on a tangent already.
I am going to go out on a limb here and propose my own approach for the desktop and not because it's the shiny new toy, or because it is hipper and can generate more buzz, but because it solves a number of problems and I cannot find money to fund the project to get it out of my dining room and its 'nights and weekends' status. Ready? Here goes:
1. Grab a copy of Ubuntu. Save yourself $200 in license fees out of the gate for Windows Vista Upgrade (Business SP1)
2. Brand it, hack it, and lock it down
3. Save the image to a server (or servers) for download
4. Link to it and email the link
5. Script an update tool so that upon login image is checked to insure there are no updates. If there are the OS grabs it, if not it stays as is.
Why Ubuntu - It's so easy to install even I can do it, and I have marginally more grey matter than a caveman. It comes with a browser, email client (although who doesn't have OWA, Gmail, Yahoo, etc now anyway), and doesn't have over 500,000 pieces of Malware, Spyware, etc. written to mess with it on a daily basis.
What do you gain - a more ecure OS, savings of $200/per desktop, no investment in RAM upgrades ($75 for 2 GB per machine), and what about agents used to keep the crap off the machine (5 agents at $100 per is $500/per machine), and support calls dealing with 'My Internet is Broken' to the 'What does a Fatal Error mean? That's not good - right?' questions. Even offshore support will set you back ~$20 call in hard costs plus lost productivity.
Net-Net you're at $700-1000 per device. You decide how much you want to save.
Gotchas - Not every app runs on Linux. Duh. That won't change and you will always have users running Windows. But you probably don't buy a new car when your tires are bald and you need to pass inspection either. Save money where you can.
That's the Desktop. How about Servers?
Server Virtualization is marching right along and has been for a few years. I think it may be eclipsed by Desktops on sheer numbers alone - I'm just not sure when. This is another Tier for Virtualization and what I see is similar to what is going on in the desktop space to some extent...
Applications were built to run on physical servers, and many application vendors haven't figured out how to certify their applications for virtualized servers. This will tap the brake pedal a few times on a server virtualization scoping exercise when you start looking at what is and is not supported for your existing apps. Also don't forget that if you virtualize 1,000 boxes and some sweet new gear, that you still have 1,000 OSes to patch and you won't save a lot if any on labor. I won't even get into licensing applications since you have sockets, CPUs, cores and everything else to factor in if you'r a software vendor. Do us a favor - figure out a per VM price and sell the heck out of it.
Then there is the Network Layer and looking at virtualizing the physical footprint of network gear. This is a third tier where things are heating up. I met with one of the Vyatta guys this week, and anyone who follows this blog knows I am pretty hopped up on this company's Linux based firewall product that will go head to head with a Cisco 7200 and they have the data to prove it. Their product runs fine on my 486 box at home. That is a lot of space, power , and cooling for my data centers, and If I am fortunate enough to run A Crossbeam Chassis, I can load Vyatta onto a blade and slap it in the chassis and go for a zero footprint increase, and measurable but not noticeable power suck.
Anyway there are a bunch of sub tiers and other points I will make, but I have a conference call to jump on... Have a great weekend!
mark @ virtualizationstuff
What struck me was that these are the same types of discussions that have happened in IT since I have been dealing with it and they center arount architecture, and more specifically eating the Elephant in Bites by using tiers in our approach.
So on the VDI or Virtual Desktop front I have had discussions with some fairly large companies about VDI plans. The bulk of them are going to Virtualize Windows and the two approaches seem to be to use Citrix/Xen or one of the new startup management tools to do this. Why Microsoft never provided a way to manage desktops better in the first place is beyond me - they figured out how to deliver a security patch every Tuesday a while ago and RDS (Remote Desktop Support) has been around a while too. Alas - I am off on a tangent already.
I am going to go out on a limb here and propose my own approach for the desktop and not because it's the shiny new toy, or because it is hipper and can generate more buzz, but because it solves a number of problems and I cannot find money to fund the project to get it out of my dining room and its 'nights and weekends' status. Ready? Here goes:
1. Grab a copy of Ubuntu. Save yourself $200 in license fees out of the gate for Windows Vista Upgrade (Business SP1)
2. Brand it, hack it, and lock it down
3. Save the image to a server (or servers) for download
4. Link to it and email the link
5. Script an update tool so that upon login image is checked to insure there are no updates. If there are the OS grabs it, if not it stays as is.
Why Ubuntu - It's so easy to install even I can do it, and I have marginally more grey matter than a caveman. It comes with a browser, email client (although who doesn't have OWA, Gmail, Yahoo, etc now anyway), and doesn't have over 500,000 pieces of Malware, Spyware, etc. written to mess with it on a daily basis.
What do you gain - a more ecure OS, savings of $200/per desktop, no investment in RAM upgrades ($75 for 2 GB per machine), and what about agents used to keep the crap off the machine (5 agents at $100 per is $500/per machine), and support calls dealing with 'My Internet is Broken' to the 'What does a Fatal Error mean? That's not good - right?' questions. Even offshore support will set you back ~$20 call in hard costs plus lost productivity.
Net-Net you're at $700-1000 per device. You decide how much you want to save.
Gotchas - Not every app runs on Linux. Duh. That won't change and you will always have users running Windows. But you probably don't buy a new car when your tires are bald and you need to pass inspection either. Save money where you can.
That's the Desktop. How about Servers?
Server Virtualization is marching right along and has been for a few years. I think it may be eclipsed by Desktops on sheer numbers alone - I'm just not sure when. This is another Tier for Virtualization and what I see is similar to what is going on in the desktop space to some extent...
Applications were built to run on physical servers, and many application vendors haven't figured out how to certify their applications for virtualized servers. This will tap the brake pedal a few times on a server virtualization scoping exercise when you start looking at what is and is not supported for your existing apps. Also don't forget that if you virtualize 1,000 boxes and some sweet new gear, that you still have 1,000 OSes to patch and you won't save a lot if any on labor. I won't even get into licensing applications since you have sockets, CPUs, cores and everything else to factor in if you'r a software vendor. Do us a favor - figure out a per VM price and sell the heck out of it.
Then there is the Network Layer and looking at virtualizing the physical footprint of network gear. This is a third tier where things are heating up. I met with one of the Vyatta guys this week, and anyone who follows this blog knows I am pretty hopped up on this company's Linux based firewall product that will go head to head with a Cisco 7200 and they have the data to prove it. Their product runs fine on my 486 box at home. That is a lot of space, power , and cooling for my data centers, and If I am fortunate enough to run A Crossbeam Chassis, I can load Vyatta onto a blade and slap it in the chassis and go for a zero footprint increase, and measurable but not noticeable power suck.
Anyway there are a bunch of sub tiers and other points I will make, but I have a conference call to jump on... Have a great weekend!
mark @ virtualizationstuff
Thursday, July 17, 2008
Observations from the VMUG New England
I am at the VMWare User Group meeting
A quick poll by me earlier and what I saw at the presentation was that there is a lot of tire kicking going on between the VMx vendors. Xen and VMWare ESX are clearly the leaders, but for different reasons - especially in the channel. VMWare was claimimg that their 3.5 ships with a lot of features that Citrix/Xen does not so from a channel view there is less money to be made for the VAR with VMWare since they bundle in a lot of stuff that you can make margin on with Citrix/Xen by augmenting the Citrix/Xen solution with other products. The downside is for the customer that if you have to license comparable products to what 3.5 ships with thats a lot of vendor relationships to manage (isn't that what the VAR's do?). Makes me wonder if VMWare's channel strategy will mimc Novell's... They need to fix that messaging in my opinion if they are to be successful in the channel and garner their channel's support, otherwise MSFT will kick the crap out of them because they figured out the channel a LOOOOONG time ago and it drives their business. StongChannel with a hot new product could gobble up market share quickly.
the next question was answered by saying 'we'd be stupid to think that VMWare won't see increased competition and some market share loss to MSFT, but we believe we are 16-18 months ahead of them in product development' - in other words, they'll get there, we're still better for the next 1-2 years, so stick with us. They also threw up some slides that I will try to get over to you if I can.
A quick poll by me earlier and what I saw at the presentation was that there is a lot of tire kicking going on between the VMx vendors. Xen and VMWare ESX are clearly the leaders, but for different reasons - especially in the channel. VMWare was claimimg that their 3.5 ships with a lot of features that Citrix/Xen does not so from a channel view there is less money to be made for the VAR with VMWare since they bundle in a lot of stuff that you can make margin on with Citrix/Xen by augmenting the Citrix/Xen solution with other products. The downside is for the customer that if you have to license comparable products to what 3.5 ships with thats a lot of vendor relationships to manage (isn't that what the VAR's do?). Makes me wonder if VMWare's channel strategy will mimc Novell's... They need to fix that messaging in my opinion if they are to be successful in the channel and garner their channel's support, otherwise MSFT will kick the crap out of them because they figured out the channel a LOOOOONG time ago and it drives their business. StongChannel with a hot new product could gobble up market share quickly.
the next question was answered by saying 'we'd be stupid to think that VMWare won't see increased competition and some market share loss to MSFT, but we believe we are 16-18 months ahead of them in product development' - in other words, they'll get there, we're still better for the next 1-2 years, so stick with us. They also threw up some slides that I will try to get over to you if I can.
Monday, July 7, 2008
What isn't being virtualized?
I hope my friends in the US had a nice long weekend. Hard to believe July 4th has come and gone...
While boogeyboarding the other evening I got to thinking about my time at Catalyst and the presentations I went to and what it meant. The sound byte I kept coming back to was - What isn't being virtualized?
Servers, desktops, and firewalls are all on a path of being virtualized at most companies. Some faster than others. What I keep wondering about is the security in all of this virtualizing and had a few thoughts/questions:
1. Is a VM akin to a VLAN - get into one and you get the keys to the kingdom?
2. Why would a company want to virtualize Windows? The same problems exists with malware, viruses, etc. and the inherent security issues.
3. Are virtual firewalls an answer or just the next new (virtualized) thing?
If I think about it, here is what I come up with:
Virtualization can create a more porous environment that breaches can exploit far easier and most likely faster.
It is the equivalent of checking into a brand new hotel and because the processes that have been in place at other properties have not been followed to excruciating discipline in a rush to open, capture excitement, etc. new holes exist, and we get a master key vs. a room key as a metaphoric example.
Why not virtualize a desktop on Linux. With close to 1,000,000 exploits out there for Windows, and only a handful for Linux - why not push Ubuntu out to a desktop and have the control you want and take 999,999,990 threats off the table at the OS?
Add a firewall (I played around with Vyatta and was impressed), and that will help, take a virtual firewall and put it in between VMs, apps, etc. and you may be on your way to taking the best practices we know and love to the virtualized world.
Thoughts? Comments?
While boogeyboarding the other evening I got to thinking about my time at Catalyst and the presentations I went to and what it meant. The sound byte I kept coming back to was - What isn't being virtualized?
Servers, desktops, and firewalls are all on a path of being virtualized at most companies. Some faster than others. What I keep wondering about is the security in all of this virtualizing and had a few thoughts/questions:
1. Is a VM akin to a VLAN - get into one and you get the keys to the kingdom?
2. Why would a company want to virtualize Windows? The same problems exists with malware, viruses, etc. and the inherent security issues.
3. Are virtual firewalls an answer or just the next new (virtualized) thing?
If I think about it, here is what I come up with:
Virtualization can create a more porous environment that breaches can exploit far easier and most likely faster.
It is the equivalent of checking into a brand new hotel and because the processes that have been in place at other properties have not been followed to excruciating discipline in a rush to open, capture excitement, etc. new holes exist, and we get a master key vs. a room key as a metaphoric example.
Why not virtualize a desktop on Linux. With close to 1,000,000 exploits out there for Windows, and only a handful for Linux - why not push Ubuntu out to a desktop and have the control you want and take 999,999,990 threats off the table at the OS?
Add a firewall (I played around with Vyatta and was impressed), and that will help, take a virtual firewall and put it in between VMs, apps, etc. and you may be on your way to taking the best practices we know and love to the virtualized world.
Thoughts? Comments?
Thursday, May 15, 2008
Virtualization – The chargeback conundrum
Funding it vs. Paying for it...
I have recently had 6 discussions in the past two weeks with companies on how to get a Virtualization project funded, which is a relatively simple discussion to have. It basically involves us working with our ROI model to come up with real cost savings on quantifiable (hard) costs. Based on that the projects are either funded or not and we help roll out infrastructure (on premise or a third party site), licensing, etc. – the usual project lifecycle stuff.
Where clients fall down in on how to get business units and departments to pay for what they use. I have dubbed it the Chargeback Conundrum.
The issue is that when an enterprise wants to execute an Enterprise License Agreement (an ELA) for the Virtualization software (VMWare) there are no simple controls or visibility into what businesses consume and how to charge them back. From the business perspective, they are reluctant to go all in to help pay for it when there is no proof of what they are getting or understanding of what they will be charged, and arguably more important – no trust with the IT organization.
In other words –
Business Executive: How do I know that I am not giving a piece of my budget to another group?
IT Executive: (insert explanation here).
Business Executive: OK, prove it.
IT Executive: (insert deer in the headlights look on face here)
If you need help – reach out – inquiries@virtualizationstuff.com
We can give you unprecedented visibility as product or a service so if you don’t want the cap ex on your balance sheet we can accommodate.
The services offering for Virtualization is also preferred when you want to pay for the functionality vs. be responsible for it.
You’ll need to work with your facilities folks to look at power consumption. While you can shrink the footprint, the power per cabinet jumps from 4KW to 7KW (almost doubles), and you’ll want to have the specialized floor tiles with variable speed fans, the temperature sensors on the front and back of the cabinets with the blades in them, and have it tied together in a management console.
If it sounds too cumbersome – reach out. We’re great at cutting to the chase…
inquiries@virtualizationstuff.com
I have recently had 6 discussions in the past two weeks with companies on how to get a Virtualization project funded, which is a relatively simple discussion to have. It basically involves us working with our ROI model to come up with real cost savings on quantifiable (hard) costs. Based on that the projects are either funded or not and we help roll out infrastructure (on premise or a third party site), licensing, etc. – the usual project lifecycle stuff.
Where clients fall down in on how to get business units and departments to pay for what they use. I have dubbed it the Chargeback Conundrum.
The issue is that when an enterprise wants to execute an Enterprise License Agreement (an ELA) for the Virtualization software (VMWare) there are no simple controls or visibility into what businesses consume and how to charge them back. From the business perspective, they are reluctant to go all in to help pay for it when there is no proof of what they are getting or understanding of what they will be charged, and arguably more important – no trust with the IT organization.
In other words –
Business Executive: How do I know that I am not giving a piece of my budget to another group?
IT Executive: (insert explanation here).
Business Executive: OK, prove it.
IT Executive: (insert deer in the headlights look on face here)
If you need help – reach out – inquiries@virtualizationstuff.com
We can give you unprecedented visibility as product or a service so if you don’t want the cap ex on your balance sheet we can accommodate.
The services offering for Virtualization is also preferred when you want to pay for the functionality vs. be responsible for it.
You’ll need to work with your facilities folks to look at power consumption. While you can shrink the footprint, the power per cabinet jumps from 4KW to 7KW (almost doubles), and you’ll want to have the specialized floor tiles with variable speed fans, the temperature sensors on the front and back of the cabinets with the blades in them, and have it tied together in a management console.
If it sounds too cumbersome – reach out. We’re great at cutting to the chase…
inquiries@virtualizationstuff.com
Tuesday, April 15, 2008
Thanks VMWare
Friday, January 11, 2008
Virtualization Hosting Dependencies
I mentioned that there were a few dependencies around virtualized hosting which in my opinion are not that dramatic, unless of course you own a data center that needs to be refurbished...
Power is the biggie, with cooling right up there as well. Most cabinets/racks draw 4KW a rack (42U cabinet) if you through a loaded Blade rack in the data center, you are looking at 7KW, essentially double. So what, right?
You will need to have the ability to do high density hosting which means 200W a square foot vs. the standard 100W a square foot.
For cooling you will need/want special tiles in the floor that increase airflow around the cabinets. For safety's sake you'll probably want to drop another CRAC in for sites with seasonal spike in utilization and corresponding heat signature(s).
If you want a green angle on the cooling, look North my friends. The smart data centers are using mother nature's inherent cooling capabilities to keep gear cool. They shut down several CRACs and use an exchanger to draw very cold dry air into the facilities, reducing the carbon footprint of that data center through reductions in power usage and cooling.
Depending upon the age of data centers this will vary from extremely difficult/expensive/not worth it to a few weeks of refitting.
Power is the biggie, with cooling right up there as well. Most cabinets/racks draw 4KW a rack (42U cabinet) if you through a loaded Blade rack in the data center, you are looking at 7KW, essentially double. So what, right?
You will need to have the ability to do high density hosting which means 200W a square foot vs. the standard 100W a square foot.
For cooling you will need/want special tiles in the floor that increase airflow around the cabinets. For safety's sake you'll probably want to drop another CRAC in for sites with seasonal spike in utilization and corresponding heat signature(s).
If you want a green angle on the cooling, look North my friends. The smart data centers are using mother nature's inherent cooling capabilities to keep gear cool. They shut down several CRACs and use an exchanger to draw very cold dry air into the facilities, reducing the carbon footprint of that data center through reductions in power usage and cooling.
Depending upon the age of data centers this will vary from extremely difficult/expensive/not worth it to a few weeks of refitting.
Thursday, January 3, 2008
Actual ROI of Virtualization
As I look ahead to 2008 and pay attention to what my customers are asking me about and what we are discussing, One of the top 2 things is virtualization which ties into the bigger picture of Infrastructure Management. The other is reducing the cost of Email management.
The top reasons companies are talking about it are:
1. Cost savings
2. It is considered green
3. It is a way to create space in a full data center
The cost savings moved the discussions along from ‘What is this virtualization thing everyone is talking about’ to ‘How do we build a plan to virtualize parts of our infrastructure?’. I will say that in my personal experience, the impacts were pretty dramatic in a data center move I was a part of. Here are the numbers:
Overall Consolidation ratio: 20:1
1,000 Wintel boxes into 50 Sun Blades running VMWare
8,000 square feet to ~200 Square feet
Cooling is 1/10 of what it was
Power is 1/10 of what it was
For a couple of large accounts that I work with, I will take you through the back of the napkin math we did on a whiteboard to quantify the ROI of Virtualization:
100,000 physical Wintel servers collapse into 5,000 Sun Blades
Power is reduced to 1/10
Cooling is reduced to 1/10
Floor space in two data centers 10:1 reduction in footprint
The assumptions were that the costs (they are leased machines) were a wash on the hardware:
• The Wintel boxes draw was 230 Watts at 50% Utilization so 23,000 KW per month
• @ $0.35/KW multiplied by 730 Hours in a month comes out to ~$6M per month on power costs to run machines and cool them.
• Did not include facilities costs
Virtualization costs:
• 5,000 blades draw 3900kw*730*.35= $900,000/mo (6 to 1 reduction)
• Add in costs of VMWare - $5,000 per instance * 5000 = $25M
• 5 month payback w/license inclusion
Are there dependencies? Yes, and in the next entry, I will explore some of the dependencies, which are negligible IMHO.
The top reasons companies are talking about it are:
1. Cost savings
2. It is considered green
3. It is a way to create space in a full data center
The cost savings moved the discussions along from ‘What is this virtualization thing everyone is talking about’ to ‘How do we build a plan to virtualize parts of our infrastructure?’. I will say that in my personal experience, the impacts were pretty dramatic in a data center move I was a part of. Here are the numbers:
Overall Consolidation ratio: 20:1
1,000 Wintel boxes into 50 Sun Blades running VMWare
8,000 square feet to ~200 Square feet
Cooling is 1/10 of what it was
Power is 1/10 of what it was
For a couple of large accounts that I work with, I will take you through the back of the napkin math we did on a whiteboard to quantify the ROI of Virtualization:
100,000 physical Wintel servers collapse into 5,000 Sun Blades
Power is reduced to 1/10
Cooling is reduced to 1/10
Floor space in two data centers 10:1 reduction in footprint
The assumptions were that the costs (they are leased machines) were a wash on the hardware:
• The Wintel boxes draw was 230 Watts at 50% Utilization so 23,000 KW per month
• @ $0.35/KW multiplied by 730 Hours in a month comes out to ~$6M per month on power costs to run machines and cool them.
• Did not include facilities costs
Virtualization costs:
• 5,000 blades draw 3900kw*730*.35= $900,000/mo (6 to 1 reduction)
• Add in costs of VMWare - $5,000 per instance * 5000 = $25M
• 5 month payback w/license inclusion
Are there dependencies? Yes, and in the next entry, I will explore some of the dependencies, which are negligible IMHO.
Thursday, December 20, 2007
Inaugural Post
Welcome to my third blog and the inaugural post.
I have decided to augment the other areas of IT I cover at IdentityStuff and PCIStuff with something that I consider to be a game changing catalyst of IT Management - Virtualization. I will explore new technologies, tried and true technology application as well as some topics related to making IT Greener.
I am working with some clients right now that are yielding some breathtaking numbers around truly quantifying the ROI of Virtualization. I will share those numbers when I am done crunching them.
Welcome, and as always, feel free to let me know if I am full of it - good or bad...
I have decided to augment the other areas of IT I cover at IdentityStuff and PCIStuff with something that I consider to be a game changing catalyst of IT Management - Virtualization. I will explore new technologies, tried and true technology application as well as some topics related to making IT Greener.
I am working with some clients right now that are yielding some breathtaking numbers around truly quantifying the ROI of Virtualization. I will share those numbers when I am done crunching them.
Welcome, and as always, feel free to let me know if I am full of it - good or bad...
Subscribe to:
Posts (Atom)
Posts
