OpenStack Components

Compiling a list of OpenStack services and components in one location is difficult. They can easily be missed because they are so widespread on the internet. Additionally, the services are not clearly stated when you do see a list of them. For your quick reference, below is a concise list that describes each OpenStack component. You should be familiar with these words because Open Stack is used worldwide. Both in private or public clouds.

The components of OpenStack are;

Compute

Nova (  Compute Service )

Pools of computational resources, including virtual machines (VMs), bare metal ( using ironic), and containers ( with limited support), are managed by Nova. One of Nova’s advantages is that it is compatible with various hypervisors, including KVM, VMware, LXC, XenServer, and others. Because of Nova’s architecture, cloud design is flexible and can be done without requiring proprietary software or hardware. It can also be used to combine third-party products and existing systems. Hypervisor technologies like KVM, VMware, LXC, XenServer, etc. can be used to deploy Nova. Diverse virtual machines as well as instances that perform different computing activities are managed by it.

Zun ( Containers Service )

Cloud customers require a more straightforward method of managing containers as greater numbers of apps are being containerized. Zun gives consumers an easy-to-use method of managing containers via OpenStack APIs without requiring them to comprehend the intricacies of containers. While Zun could handle containers without relying on the Nova API, Nova requires a Nova docker driver. 

Hardware Lifecycle

Ironic ( Bare Metal Provisioning)

OpenStack often provides Virtual Machines ( VMs) for provisioning, but bare metal provisioning may be necessary at times. This is made possible by this service by utilizing OpenStack APIs. It also interfered with Nova, allowing the latter to offer bare metals using its services.

Both standard remote administration protocols ( such as PXE and IPMI) and vendor-specific ones are used by the bare Metal service to manage devices. In addition to giving the computer service an interface that lets physical servers be handled like virtual machines, it gives the cloud operator a single access to a diverse fleet of servers.

Cyborg ( Accelerator)

Accelerators (GPU, FPGA, ASIC, Np, So’s, NVMe/ NOF SSD, ODP, DPDK/SPDK) are necessary for Telco workloads. Cyborg uses the API either directly or via utilities to develop and oversee the accelerators.  

You can oversee the acceleration lifespan of a system in a cloud computing environment by using the OpenStack Cyborg acceleration service. It makes controlling the accelerators connected to instances simple. 

Storage

Cinder( Block Storage) 

You may give your virtual machines long-lasting storage by using Cinder, an OpenStack Block storage service. It works in conjunction with OpenStack Compute to cope with volume kinds and snapshots, as well as to supply volumes for instances.

Cinder is compatible with Linux servers, Hitachi data systems, EMC ( ScaleIO, VMAX, and VNX), Ceph, Coraid, CloudByte, IBM, and SAN volume controllers, among other storage platforms. It works well for data storage and scalable file systems.

Manila ( Shared file system )

Manila offers communal storage. Manila storage permits for sharing across several VMs and omstsnxes, whereas Cinder only offers it for a single VM. Manila, which was inspired by the Cinder project, offers a legitimate storage distribution control plane in OpenStack for decentralized or shared file systems, just like Cinder dies for block storage.

By enabling external storage systems to be mapped to Nova hosts and guests via file-based NAS protocols, Manila fills the discrepancy across block and object. As the NAS protocol handles the locking and data integrity procedures needed to enable numerous compatible access to data, file shares can be shared between hosts and guests.  

Swift ( Object storage)

Have you heard of AWs’s S3? It’s an affordable solution for storing enormous amounts of data. Swift offers the same kind of assistance. Another name for it is Object storage. It is widely dispersed and offers an affordable scale-out object storage solution. 

Swift is an object storage that is highly accessible, shareable, and eventually consistent. It makes large amounts of data secure, affordable, and effective to store for organizations. Swift’s ability to guarantee data distribution and replication across several devices makes it the perfect choice for scale-out, affordable storage. 

Orchestration ( Heat)

With the Havana release, the Orchestration project Heat makes its debut. Heats early development was restricted to a small number of OpenStack resources, such as network, compute, image, and block storage services. By coordinating various cloud resources, Heat has accelerated the development of resource management OpenStack, enabling the design of stacks that can launch applications with a few button clicks. Users can quickly provision the needed resources and launch programs from template engine text files known as HOT templates ( Heat Orchestration Template).