Cloud Technical Glossary
Jump to navigation
Jump to search
Term | Description |
---|---|
Apache HTTP Server | A Web server software. See https://en.wikipedia.org/wiki/Apache_HTTP_Server. |
CALM (cloud account lifecycle management) | Our process for managing the allocation of cloud resources. |
Ceph | The distributed data storage platform for our clouds; includes volume storage. |
CephFS (Ceph File System) | A filesystem with Ceph storage, which allows data to be mounted simultaneously on multiple OpenStack instances. Currently only available with Arbutus. See https://docs.ceph.com/en/latest/cephfs/. |
cloud | When referring to our cloud services, short form of IaaS cloud. |
compute cloud | Type of resource allocated to support OpenStack instances intended to run for a limited time, usually with a very high sustained usage of CPU and memory. Flavours for these resources have labels beginning with c*. Currently only available on Arbutus, Béluga and Cedar. Compare with persistent cloud. |
CVMFS (CernVM File System) | A read-only content distribution system often used to manage software. See https://docs.alliancecan.ca/wiki/CVMFS. |
ephemeral local disk | Virtual disk which is created and destroyed along with an OpenStack instance. An ephemeral disk is created when an instance is launched without the specification of a volume. |
flavor | OpenStack term for a predefined specification of a new instance. A flavor can define sizes for RAM, disk, number of cores, and so on. |
floating IP | Internet Protocol address that can be associated with an OpenStack instance to allow external access. |
Horizon | OpenStack cloud dashboard that is used for viewing and managing cloud resources through a Web browser. See https://docs.openstack.org/horizon/latest/. |
host | Physical server supporting virtual machines. |
image | Image of a virtual disk used to create a new boot volume or ephemeral disk when creating an OpenStack instance. |
instance | OpenStack virtual machines are called instances, mostly because they are instances of an image that is created upon request and that is configured when the instance is launched. |
IPV6 (Internet Protocol version 6) | A communications protocol successor to IPv4. See https://en.wikipedia.org/wiki/IPv6. |
object storage | Object storage (or object-based storage) is a storage type that manages data as objects, as opposed to other storage architectures like filesystems which manage data as a file hierarchy, and storage types where data is managed as blocks. Each object typically includes the data itself, a variable amount of metadata, and a globally unique identifier. Offered as S3 and Swift protocols. Allocated in TB. See https://en.wikipedia.org/wiki/Object_storage. Currently only available with Arbutus. |
OpenStack | The software suite used on our clouds to control hardware resources such as computers, storage and networking. |
persistent cloud | Allocation type for persistent virtual machines that are expected to run indefinitely and have low or bursty CPU requirements. Flavours for these resources have labels beginning with p*. Currently only available on Arbutus, Béluga and Cedar. Compare with compute cloud. |
project | In our infrastructure, a project represents an allocation of cloud resources to a group or user. Sometimes referred to as tenant. |
RAC (Resource Allocation Competition) | Our program via which PIs can submit a request for storage and computer resources beyond what can be obtained via the Rapid Access Service (RAS). The requests are evaluated by a committee of peers. |
RAS (Rapid Access Service) | Service by which Principal Investigators can request a modest amount of storage and cloud resources without having to apply to the RAC. |
S3 (Simple Storage Service) | A type of object storage. See https://en.wikipedia.org/wiki/Amazon_S3. |
security group | A set of security rules that control network traffic and can be applied as a whole to one or more instances. |
service portal | Our infrastructure hosts many Web research portals which serve datasets or tools to a broad research community. These portals generally do not require large computing or storage resources, but may require support effort by our technical team. Groups applying for a service portal often use our clouds, generally require a public IP address, and may have more stringent up-time requirements than most research projects. |
shared filesystem | Persistent storage space offered as a Unix-compliant filesystem that can be mounted across multiple hosts in a project. This is useful for sharing data across multiple hosts. Service runs on CephFS and requires either a Fuse driver (Windows/Linux) or the CephFS kernel driver (Linux) for access. Allocated in TB. |
snapshot | Copy of an OpenStack volume that can be used as a backup or to launch another instance. |
SSL (Secure Sockets Layer) | A protocol to allow encrypted communications over networks. SSL is obsolete and should be replaced by TLS (Transport Layer Security) wherever possible. |
SWIFT | A type of object storage. See https://wiki.openstack.org/wiki/Swift. |
tenant | See project. |
TSL (Transport Layer Security) | See SSL. |
vCPU (virtual central processing unit) | A vCPU represents a portion or share of the underlying, physical CPU that is assigned to a particular virtual machine. |
vGPU (virtual graphics processing unit) | A virtual machine can have one or more vGPUs assigned to it. Each of them is seen as a GPU by the operating system. Additional configuration may be required for use. See https://docs.alliancecan.ca/wiki/Using_cloud_vGPUs. |
virtual machine | Virtual server in the cloud infrastructure. In OpenStack, active virtual machines are called instances. |
volume | Storage resource that can be attached to or detached from an OpenStack instance, like a virtual disk. |
volume storage | Type of persistent cloud storage providing virtual disk functionality to OpenStack instances running in the cloud. Implemented with Ceph software. Allocated in GB. |