Kioxia KumoScale V.3.14 Storage Software Based on NVMe-oF
Delivers enhanced utilization, capacity management and reliability to networked NVMe flash at data center scale.
This is a Press Release edited by StorageNewsletter.com on July 3, 2020 at 2:26 pmKioxia America, Inc. introduced a version of its KumoScale storage software based on NVMe-oF technology.
With the addition of features designed to support cloud providers, the software suite enables next-gen, cost effective, private cloud deployments.
Developed to address the heightened demand for NVMe flash storage and the NVMe-oF framework that makes cloud-based disaggregation a reality, the software is a cloud-focused solution that delivers performance NVMe flash storage volumes via a virtual storage layer between clients and flash devices. It provides a central API through which applications can request the resources needed, then dynamically provision and connect the right amount of shared flash storage for each application instance. KumoScale software is NVMe-oF compliant and leverages the spec to deliver performance as if the NVMe SSDs were locally attached, and is one of the few storage fabric solutions that supports bare-metal, virtualized and Kubernetes deployments on a range of industry-standard servers.
Enhancements to V3.14 include:
-
Thin provisioning: It allows storage resources to be allocated to servers, on a just-enough and just-in-time basis. KumoScale software’s thin-provisioned volumes allocate blocks of data on demand vs. the traditional method of allocating all possible blocks to a client in advance. This eliminates almost all whitespace and increases storage utilization rates, which can often be as low as 10% using the traditional approach.
-
Autonomous self-healing: KumoScale software’s volume self-healing mechanism automatically identifies and repairs data inconsistencies that may result from replica failures, including reconnection to the target or an incomplete replica deletion, without the need for monitoring and intervention by a system administrator.
-
Network resilience: Software clients use Linux Link Aggregation Control Protocol (LACP) and enhanced connections management to ensure consistent delivery of packets across all available network paths. It targets use port bonding to ensure availability while maximizing total storage node throughput.
-
Lights-out storage node deployment: Preboot eXecution Environment (PXE) protocol support enables automated installation over a network in data center environments. PXE boot installation capability replaces the need for either a bootable device for the OS or the need to use an ISO file on a DVD or USB device.
-
Snapshots: Snapshots for Kubernetes deployments via the Kubernetes CSI API, and directly via the REST API or CLI.
-
Support for performance CM6 Series SSDs: Support for the firm’s CM6 Series of PCIe 4.0 ((1×4, 2×2) NVMe 1.4 enterprise SSDs. CM6 drives deliver sequential and random performance of up to 6.9GB/s and up to 1.4 million IO/s, benefiting applications such as databases, data analytics and AI.
“KumoScale 3.14 rolls up an impressive list of advanced storage functionality and demonstrates our continued commitment to automating the deployment and management of NVMe flash storage, making it easier to adopt and less expensive to use,” said Joel Dedrick, VP and GM, networked storage software. “We’re delivering on our promise to help make high-performance NVMe flash the clear choice for mainstream data center storage.”
Read also:
Kioxia KumoScale Software Enhances Cloud Storage Resilience With Support for Autonomous Self-Healing
Enables automated recovery when replicated volumes become unavailable.
June 26, 2020 | Press Release
Kioxia Adds Thin Provisioning to KumoScale Software Suite Based on NVMe-oF
Maximizes flash utilization for disaggregated cloud storage while reducing costs.
June 5, 2020 | Press Release
Kioxia Enhanced KumoScale Storage Software Based on NVMe-oF to Include Kubernetes CSI-Compliant Snapshots and Clones
Addition of storage functions enables software to reduce time pressures associated with backup processes for large, stateful applications such as databases – while consuming little storage space.
May 1, 2020 | Press Release