Elastic Compute provides a virtualization environment based on CloudStack as management middleware and KVM as hypervisor. The platform is configured, set up, maintained, and monitored by Leaseweb.
Customers can get their resources (such as CPU/RAM and storage) on this fully redundant platform. Elastic Compute customers get access to their resources using the native CloudStack Graphical User Interface and API. This enables them to utilize their resources to full extent using the CloudStack management suite.
Elastic compute comes with two types of resource options for different performance needs:
Elastic Compute for a more competitively priced option without sacrificing excellent computing performance. This option is currently only available in AMS-01.
Elastic Compute Premium has dedicated hyperthreads ready to maximize performance and handle your most demanding workloads. This option is currently available in AMS-01, LON-01, SIN-11, FRA-10, SFO-12, and WDC-02.
Leaseweb supplies some basic templates and ISOs to get you started. However, Leaseweb does not supply more than these basic templates and ISOs.
The Private Cloud enables you to upload and create your own templates and ISOs, thereby giving full flexibility to you in regards to OS choice.
Leaseweb does not supply (pre-licensed) windows OS templates.
We do supply prepared Windows ISOs so that you can start installing Windows yourself. The licenses for Windows can be acquired through us (please contact Sales) and are not included in default pricing. It is also possible to use your own licenses with this ISO.
Control panel software
We do not supply pre-created templates with control panel software. However, you can buy licenses and install any panel.
Cloudstack provides a wealth of options with respect to networks. The two basic network models available in Leaseweb Private Cloud are:
- Shared networks (instances get a public IPv4 address directly assigned to their virtual NIC)
- Isolated networks (all instances are deployed behind a virtual router device within a private network using private IP space)
Particularly the isolated network provides a lot of features including:
- Loadbalancers (basic load balancing by the virtual router device for an isolated network)
- Firewalling (both ingres and egres)
- Port forwarding (through the virtual router to specific instances within the virtual routers isolated network)
- Possibility to create a Virtual Private Cloud, VPC (a network consisting of multiple isolated networks, tiers, all connected and managed through the same virtual router)
- Possibility to create site to site VPNs (with a VPC)
Within Elastic Compute, there is a firewall functionality for virtual private networks. This functionality is delivered by means of the virtual router and is part of the default product. It does not incur any additional cost.
For Private Cloud, it is possible to buy ranges of public IPv4 addresses starting with a /29. These IP addresses are charged per IP address.
You need to acquire these ranges per network model you require. Therefore, if both the shared and the isolated networks are needed, then two ranges are required to give those connectivity to the Internet. Please keep in mind that due to the nature of subnets, the effectively usable IP addresses are less than the total number bought. Regular procedures for obtaining these addresses apply. If more than a /28 range is required, it is possible to get them but regular procedure for obtaining those IP addresses apply. Sales can help you with that request.
It is possible to obtain SLAs for Private Cloud based on Cloudstack. This will ensure fast support and a quicker response time.
Upgrades and Downgrades
You can upgrade and downgrade your Private Cloud (increase or decrease your resources) by contacting Support. Depending on the required resources, most upgrade requests can be fulfilled in hours.
Delivery time depends on the size of the required resources (cores, storage, IP addresses). Most requests can be delivered within hours.
CloudStack Private Cloud APIs
The Private Cloud API will provide you the ability to retrieve information about the Private Clouds that you have at Leaseweb. Using APIs, you will be able to list or retrieve your Private Clouds, list them, access credentials, get all types of metrics like datatraffic, CPU, memory and storage.
Check out the complete list of Private Cloud APIs.
Cloud Bursting with Elastic Compute
As an Elastic Compute customer with a committed resource pool, you can scale your resources up and down on an hourly basis in case of periodic or seasonal needs. CPUs and RAM can be scaled separately to accommodate fluctuating workloads. Obtain financial advantages with a flat fee pricing structure and eliminate the need to commit to resources peaks. Bursting is only available in DE, NL, and UK entity.
When is cloud bursting needed?
- For periodic resource needs: during a period of new deployment, a new product launch, a marketing campaign, etc.
- During rapid unpredictable growth period: to see what would be needed before committing to a larger pool.
- To cover unpredictable jumps in workloads.
Enabling cloud bursting with Elastic Compute
Perform the following steps to enable cloud bursting:
- Log in to the Customer Portal and click on Private Cloud.
- Under Elastic Compute, choose the resource pool that you want to enable bursting for.
- Click Manage Bursting under Actions.
- Activate bursting by switching the Bursting button to Enabled, and then click Submit.
- When the committed resource pool capacity is full, new virtual machines (VMs) will be created with bursting and will be charged by the hour. This happens separately for CPUs and RAM.
The new limit is set as 1.5 times the committed resource pool. This cannot be changed.
Invoicing process for cloud bursting
When you will be invoiced for cloud bursting, these factors shall be considered:
- A flat fee contract is invoiced at the start of the month.
- The actual usage for each item is reported on an hourly basis.
- In case an item (CPU, RAM or both) is used over the committed resource pool, you will be charged hourly for the item used. Over-usage calculations are as follows:
CPU and RAM: unit price for the item x number of hours used x amount that is overused
Note: Unit prices are on demand list prices.
Reporting on cloud bursting
You can use the reporting feature to do the following tasks:
- See your real usage in “Usage Graphs” on the Customer Portal.
- Graph shows the usage of storage, CPUs, memory allocated to VMs, and data traffic usage.
- Next to real usage, flat fee resource pool amounts are displayed along with the over-usage amounts.
Upgrading resource pools for cloud bursting
If you have cloud bursting enabled, in case of a resource pool upgrade request, your new bursting limit will also upgrade at 1.5 and your bursting feature will stay enabled.
Disabling cloud bursting
Perform the following steps to disable cloud bursting:
- Login to the Customer Portal and click on Private Cloud.
- Under Elastic Compute, choose the resource pool that you want to disable bursting for.
- Click Manage Bursting under Actions.
- Disable bursting by switching the Bursting button to Disabled, then click Submit.
- Leaseweb will verify that the actual usage does not exceed the resources pool flat fee amount.
- If the actual usage exceeds the flat fee resources available from your contract, a message will be shown saying that over used resources needs to be cleaned up before bursting can be disabled.
Leaseweb provided documentation
We have written an extensive manual explaining the available functions and tasks you can perform in the Elastic Compute.
- Getting Started with Apache CloudStack
- Viewing Apache CloudStack dashboard
- Managing Apache CloudStack Instances
- Managing Apache CloudStack Affinity Groups
- Managing Apache CloudStack Storage
- Managing Apache CloudStack Network
- Managing Apache CloudStack Templates
- Deleting Apache CloudStack Events
- Managing Apache CloudStack Projects
- Managing Apache CloudStack Accounts
- Managing Apache CloudStack Domains
- Managing Apache CloudStack Service Offerings
- Managing Apache CloudStack API and automation
Cloudstack provided documentation
There are a lot of generic documentation explaining the infrastructure setup and generic concepts within CloudStack. Particularly the "Concepts and Terminology" and "Programmer guide" can be of great use.
The first gives a basic understanding of Cloudstack setup and working. The second is a great source for getting started with the API.
There is the Administrators guide which provides a lot of insight in the setup and workings of CloudStack:
The Release Notes will tell you what has changed between the versions. This can be useful in case our Private Cloud version is not yet updated to the latest one (which will always be a temporary situation):
There is the API guide which will give insight in the available API calls. We give our customers domain admin access, and hence the Domain Admin guide should be used. Together with the "Programmer guide", this provides a good starting point for using the API:
There is a wiki which contains additional information: