Scribd Logo
Upload

Welcome to Scribd!

  • 8 views

    Container Native Storage For DevOps

    Uploaded by

    davidsod

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Container Native Storage For DevOps

    Uploaded by

    davidsod
    8 views2 pages

    Original Title

    Container-Native-Storage-for-DevOps

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    8 views2 pages

    Container Native Storage For DevOps

    Uploaded by

    davidsod

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 2

    SOLUTION BRIEF

    Container-Native Storage for DevOps


    Dynamically provision high-performance persistent storage
    to stateful applications on Kubernetes clusters

    Modern cloud-native applications require modern containerized solutions


    to run and manage them. Traditional monolithic storage alternatives cannot
    BENEFITS address the dynamic provisioning requirements of microservices at scale, nor
    the responsiveness and location independence expected of them. DevOps
    • Accelerate responsiveness of personnel and site reliability engineers (SRE) depend on a fast, reliable,
    stateful workloads through low- and flexible infrastructure to streamline CI/CD operations and deliver high-
    latency access to NVMe drives performing applications under varied conditions.

    While Kubernetes provides the orchestration platform to automate and


    • Streamline DevOps CI/CD manage the container lifecycle, many projects underestimate critical
    pipeline with containerized dependencies on the data storage layer, especially in terms of performance,
    storage services orchestrated integration, and portability. Storage related problems that are not apparent
    during proof of concept become evidently clear in production when spread
    directly from Kubernetes out over hundreds or thousands of pods in diverse environments.

    • Replace complex storage Many of those unforeseen problems stem from relying on outside control
    of storage resources from systems and processes designed for much more
    management with native
    static and centralized operations.
    Kubernetes automation

    Ensure Responsiveness, Manageability, and Portability of


    • Gain observability to storage
    Critical Storage Services for Consistent Behavior Across
    behavior not possible with
    Diverse Container Deployments
    external systems
    A sure way to avoid external storage dependencies and their undesired
    • Ensure portability of storage surprises is by letting Kubernetes manage storage allocation, load balancing
    and failover as an integral part of maintaining your desired state. For that you
    services by removing kernel/OS
    must first containerize the most sought-after storage services. DataCore Bolt
    dependencies software offers such a solution, borne out of our extensive experience with the
    top open-source storage package, OpenEBS.
    • Enhance data resilience
    No need to be concerned with special OS or kernel calls. Bolt runs completely
    though volume replication
    in user space, ensuring widespread portability. The code you build and test
    across K8s nodes in one environment will behave the same way in another. Deploy with equal
    confidence in multi-cloud scenarios, on-premises, and at the edge.

    Bolt abstracting
    Stateful Application Stateful Application Stateful Application
    data services
    from storage and
    enabling dynamic
    volume provisioning
    Pod Pod Pod

    Kubernetes Control Plane

    Bare Metal Storage Virtualized Storage Cloud Storage

    The Authority on Software-Defined Storage datacore.com


    SOLUTION BRIEF

    You will enjoy dynamic provisioning Kubernetes Cluster

    of persistent volumes for your


    stateful workloads, that also happen Node C Node A Node B

    to be super-fast. Fully exploit the Application


    Pod
    raw speed and low latency of NVMe NVMe-oF/TCP
    SSD hardware by eliminating much Persistent
    of the overhead and interrupts that Volume (PV)

    other alternatives introduce.


    Volume Storage
    Controller
    Just deploy the lightweight Bolt NVMe-oF/TCP NVMe-oF/TCP

    container-native storage software Remote Replica Local Replica Remote Replica


    on your Kubernetes nodes – either
    bare metal or virtual servers. Bolt
    abstracts the disk attached to the Pool Pool Pool

    node and creates a storage pool.


    Use conventional declarations to
    claim persistent volumes which are Disk Disk Disk

    logically partitioned and provisioned


    to pods just like any other resource. Replication of volumes across nodes

    Leverage volume replicas within the


    cluster to protect against hardware Bolt protecting volumes against failure by creating replicas across K8s nodes
    failures and data loss.

    These simple steps facilitate the deployment and release of production-ready software while enabling enterprise-scale
    agility. Minimize wait time and accelerate code progression to meet your goals of continuous integration, delivery, and
    deployment. Harnessing the benefits of NVMe and NVMe-oF puts you in a position to support a composable and
    disaggregated storage architecture that delivers low-latency and enhanced responsiveness for stateful workloads.

    No storage expertise is needed to deploy and manage Bolt: your DevOps engineer/SRE or K8s admin can focus on
    their core tasks while Bolt provides platform-independent storage services to fulfill the needs of containers. Simply
    schedule Bolt pods to run on the same worker nodes as the application, drawing capacity from a local disk to create
    persistent volumes. Those logical volumes will be dynamically provisioned on-demand to stateful workloads and may
    be replicated to other nodes for added resilience and throughput. NVMe over TCP is used for high-speed, synchronous
    replication between nodes. The type of disk varies depending on whether the node is in the cloud, on a virtual machine,
    or on a bare metal server.

    Why Choose DataCore Bolt for Your Kubernetes Environment

    Persistent High Easy Capacity Circumvent Platform-Agnostic Software-Defined


    Volumes for Performance for Scaling Storage Outages Storage Automation
    Kubernetes Stateful Apps & Failures

    You can seamlessly snap Bolt into your DevOps ecosystem alongside other tools used for version control, configuration
    management, CI/CD automation, code repository and artifact management, test automation, and so on. Out of the
    box integration with monitoring tools (such as Prometheus) provides logs and metrics for complete observability.

    Enhance the experience and eff iciency of I/O-intensive applications using Bolt. Empower your organization
    with the agility to scale the Kubernetes inf rastructure to support stateful containerized applications and the
    DevOps pipeline. Contact DataCore to learn more about Bolt container-native storage.

    0422

    Discover the Ultimate Flexibility of DataCore Software


    DataCore Software delivers the industry’s most flexible, intelligent, and powerful
    software-defined storage solutions for block, file, and object storage, helping
    more than 10,000 customers worldwide modernize how they store, protect, and
    access data. With a comprehensive product suite, intellectual property portfolio,
    GET STARTED and unrivaled experience in storage virtualization and advanced data services,
    DataCore is The Authority on Software-Defined Storage. www.datacore.com

    © 2022 DataCore Software Corporation. All Rights Reserved. DataCore and the DataCore logo are trademarks or registered trademarks of DataCore Software Corporation. Other
    DataCore product or service names or logos referenced herein are trademarks of DataCore Software Corporation. All other products, services and company names mentioned
    herein may be trademarks of their respective owners.

    You might also like