Zesty announced the release of Zesty
Disk for K8s. This dynamic block storage
auto-scaler can now be applied to K8s clusters to right-size EBS volumes,
improve application stability, and reduce storage costs by up to 70% - all
while relieving engineers of the burden of managing volumes.
The
functionality of K8s, an open-source container orchestration tool,
significantly eases the process of provisioning EBS volumes. Ordering a
specified disk size for containers is a challenge. The process, which involves
specifying the GB size, requires constantly monitoring the volume to ensure
there is enough space at any given moment, otherwise, the stability of running
workloads may be compromised. This challenge is further compounded in cloud
environments that have hundreds or thousands of running pods, as is typical for
many applications.
K8s,
which are often used to orchestrate very large workloads associated with
databases, ETLs, and Machine Learning pipelines, have high data ingestion
requirements that can unexpectedly overwhelm standard EBS volumes. When this
happens, an application may slow down dramatically or crash due to an "out of
disk error" - both significantly impacting the user experience.
Zesty
Disk is ideal for any K8 workload, as its ability to scale storage volume means
it can be applied to both relatively simple tasks as well as workloads that
involve intense and dynamic disk usage. DevOps engineers like to use templates
as a best practice to deploy, manage, and duplicate environments in order to
standardize provisioning for infrastructure. However, each environment is
different; it is unlikely that the disk size in one development environment
will also be needed in production. Furthermore, DevOps is never linear, and as
the technology implemented evolves, changes are constantly required, making it
necessary to regularly monitor and right-size disk volumes.
By
utilizing Zesty Disk's enhanced EBS storage, volumes will automatically expand
and shrink according to real-time application needs, eliminating the need to
manually adjust or predict volume size. Automatic expansion of volume ensures
that when a huge data influx occurs, it won't impact the container's stability
or lead to downtime for an app. Conversely, automatic reduction of volume saves
a significant amount of money and other resources as it eliminates the need to
over-provision excess file system capacity and provides the unique ability to
downscale following a large data influx.
Zesty's
solution is integrated into a building block of K8s called Persistent Volumes.
Volume elasticity is enabled by decoupling large file system volumes into a
series of multiple volumes that are added or removed on the fly without
interrupting the running pod.