5 ms·

So besides the temporary discount, what does this offer that's significantly different from AWS? For example the pricing and naming scheme of HP Cloud Compute a

by pjin 15y ago

So besides the temporary discount, what does this offer that's significantly different from AWS? For example the pricing and naming scheme of HP Cloud Compute and EC2 are more or less identical, and the machine specs are on average the same (one vendor's instance has more RAM, one has more disk, etc.).

samarudge 15y ago
Name Tell an enterprise you want Amazon to host their software then spend a few days/weeks/months explaining that Amazon does more than just selling stuff. HP is well known, and trusted. For your average start-up, no, there's no difference. But for the enterprise having a name you know is very important. It's why people will pay so much for Oracle software over open source with support contracts.
- gabrtv 15y ago
  Most enterprise CIO's understand Amazon's place in cloud computing. You'd have to be criminally negligent not to. Most enterprise tools vendors are positioning for AWS support/compatibility as well. I'm attending a BMC webinar in a few minutes that's squarely focused on AWS. The point about brand names is more about loyalty, in my experience. Companies that buy Oracle will continue to buy Oracle. Companies that buy HP are more likely to pursue an HP Cloud offering. It's all about established relationships and sales channels.
musket 15y ago
You could split your service, half hosted on Amazon and the other half on the HP cloud, so next time one of them fails you can still operate..
- justinsb 15y ago
  Or half-hosted on HP, half hosted on Rackspace's OpenStack product (when it is released), so that you don't need to code to two different APIs. Or half-hosted on IBM, half-hosted on Internap, half-hosted on Dell public cloud, half-hosted in a private cloud... all with the same OpenStack API. If you decide you want to run your own hardware (as Zynga did & Justin.TV did when EC2 got too expensive), you just install OpenStack on your hardware - it is open source. This is why Amazon scrambled to partner with Eucalyptus recently, so that they could have something to point to as an alternative that supports the EC2 API.
  - viraptor 15y ago
    Actually unless you're doing something very tightly connected to one single service, there's a number of libraries you can use which will give you a nice abstraction. There was also some reverse-proxy which "normalises" your requests for a given provider presented on Fossdem this year, but I can't recall the name.
    - justinsb 15y ago
      Abstraction layers are nice in principle, but the issue isn't really the APIs, it is the different semantics. To give a concrete example, on Rackspace's (current) IaaS offering disks are local to the machine and persistent; on Amazon's offering disks are local and ephemeral, or locked to one AZ and persistent. The API issue is a triviality compared to hiding those details. I think that's the real power of OpenStack - the shared API is a convenience, but the shared model is priceless.

by pjin 15y ago

samarudge 15y ago

Name Tell an enterprise you want Amazon to host their software then spend a few days/weeks/months explaining that Amazon does more than just selling stuff. HP is well known, and trusted. For your average start-up, no, there's no difference. But for the enterprise having a name you know is very important. It's why people will pay so much for Oracle software over open source with support contracts.

gabrtv 15y ago
Most enterprise CIO's understand Amazon's place in cloud computing. You'd have to be criminally negligent not to. Most enterprise tools vendors are positioning for AWS support/compatibility as well. I'm attending a BMC webinar in a few minutes that's squarely focused on AWS. The point about brand names is more about loyalty, in my experience. Companies that buy Oracle will continue to buy Oracle. Companies that buy HP are more likely to pursue an HP Cloud offering. It's all about established relationships and sales channels.

musket 15y ago

You could split your service, half hosted on Amazon and the other half on the HP cloud, so next time one of them fails you can still operate..

justinsb 15y ago
Or half-hosted on HP, half hosted on Rackspace's OpenStack product (when it is released), so that you don't need to code to two different APIs. Or half-hosted on IBM, half-hosted on Internap, half-hosted on Dell public cloud, half-hosted in a private cloud... all with the same OpenStack API. If you decide you want to run your own hardware (as Zynga did & Justin.TV did when EC2 got too expensive), you just install OpenStack on your hardware - it is open source. This is why Amazon scrambled to partner with Eucalyptus recently, so that they could have something to point to as an alternative that supports the EC2 API.
- viraptor 15y ago
  Actually unless you're doing something very tightly connected to one single service, there's a number of libraries you can use which will give you a nice abstraction. There was also some reverse-proxy which "normalises" your requests for a given provider presented on Fossdem this year, but I can't recall the name.
  - justinsb 15y ago
    Abstraction layers are nice in principle, but the issue isn't really the APIs, it is the different semantics. To give a concrete example, on Rackspace's (current) IaaS offering disks are local to the machine and persistent; on Amazon's offering disks are local and ephemeral, or locked to one AZ and persistent. The API issue is a triviality compared to hiding those details. I think that's the real power of OpenStack - the shared API is a convenience, but the shared model is priceless.