All Blogs

100% Data Availability With Hitachi Virtual Storage Platforms

Hu Yoshida Hu Yoshida
CTO Emeritus, Hitachi Vantara

November 15, 2021

Data availability is a top concern for CIOs, especially for their storage systems. While processor and network availability are also important, servers and networks are essentially stateless and can be replaced, restarting your data processing. If you have an outage in your storage systems, it may take hours or even days to recover the state of the data.

Hitachi Vantara products are designed for reliability, and that’s why we are able to address this concern for storage data with a 100% data availability guarantee on our Hitachi Virtual Storage Platforms (VSPs). We have been offering this for over 20 years!

Our 100% data availability guarantee for our Hitachi Virtual Storage Platform family ensures that customers meet their information availability requirements. If there is a loss of availability of data caused by a malfunction of the storage system that makes it unable to read or write data, Hitachi Vantara will credit the customer for replacement equipment. All our VSP systems offer this guarantee.

Industry analysts, partners, and customers consider Hitachi storage platforms as the gold standard in quality, availability, and durability. We have customers who have never experienced an outage on any of their Hitachi enterprise storage systems. Even our harshest critics who say that there is no real difference between any of the storage vendor products, must recognize that we are the only storage vendor who has been offering a 100% data availability guarantee of this kind for this long. There are now a few vendors that offer a data availability guarantee, but that guarantee is limited to certain products and may require an expensive remote duplicate system.

Why is Hitachi Vantara the only storage vendor that can provide this guarantee across its VSP midrange and enterprise product line?

I recently reached out to the Hitachi storage hardware and software engineers and our customer services and support organizations to understand what made our products different.

With full access to the Hitachi Remote Ops monitoring system database, a random sampling of 150 million operating hours of Hitachi VSP data was gathered to better understand how we were delivering on the quality that our customer expected. We could see that over this period 40 trillion data points were being collected. In the same time period, 90% of problems were proactively identified, enabling our customer support to inform the system administrator ahead of time that service was needed prior to the customer noticing any issue.

The basic architecture of the Virtual Storage Platform hardware is unique. It is built around a switch architecture, which separates control from data so that data flows can be monitored and reported on without interference or interruption to the data flow. This is also a security feature since Hitachi service personnel can work on the system and never see the customer data. The switch capability can also enable the system to switch around failures and continue running. All hardware components are engineered to have no single point of failure with multiple levels of redundancy along with the provision of data for alerts.

The operating system for all Hitachi storage systems, from midrange to enterprise, is Hitachi Storage Virtualization Operating System RF (SVOS RF), so we are able to offer this 100% data availability guarantee across our product line. Other vendors have three or four product lines and cannot offer this type of guarantee. The VSP line provides predictive monitoring via Hitachi Remote Ops, which was built over time within the close partnership of hardware and software engineering and our services organization.

There are over 28 years of experience and more than 4,000 industry-leading patents in system engineering and support embedded in every Hitachi storage system. SVOS RF’s ongoing data collection and “phone home” reporting capability ensure quality is maintained over time by monitoring hardware, software, environmental and data path metrics to proactively ensure the system is operating under ideal conditions. On a weekly basis, Hitachi Remote Ops monitoring system collects more than 40 trillion data points to track system quality of service, enabling quick root cause analysis of any response time abnormalities. Annually, Hitachi Remote Ops monitoring system’s smart, predictive platform analyzes over 100 billion flash and hard drive information events, to ensure data integrity and availability, and to identify quality issues before they impact customers.

As new technologies like microservices, FPGA, PCIe and NVMe became available, SVOS RF was redesigned to take advantage of these new capabilities:

  • Hitachi developers reimagined core systems as microservices.
  • FPGAs offload eliminated the need for expensive custom ASICs.
  • Our proprietary switches were replaced by PCI switches for lower costs.
  • SVOS RF leveraged the massive parallelism of NVME for performance and availability.

SVOS RF also includes AI and machine learning (ML) to provide cognitive insights, which extend further than the system components, storage devices, software or firmware. It also monitors interoperability between the software function across logical and physical data path, to the network and the host. Application resilience can only be achieved with visibility of the entire stack from the host down. Hitachi’s VSP systems also support a RESTful API so that this information can be shared with other monitoring systems, like ServiceNow.

Based on this information and the experience of Hitachi customers over the past 28 years, Hitachi Vantara can go all in on 100% data availability for our current and future customers. For a company to qualify for this 100% data availability guarantee, Hitachi does require that the following availability features are in place:

  • Dual-parity RAID groups configured for data storage.
  • Redundant paths from host to SAN switches to storage ports.
  • Hitachi qualified host failover software on connected servers.
  • Hitachi qualified connected components, such as host bus adapters (HBAs), switches and other SAN or LAN components, listed on the interoperability matrix.
  • Hitachi Remote Ops monitoring system and call home feature installed and operating.

For more information, see our 100% availability datasheet or contact your Hitachi Vantara sales representative or Hitachi partner.

Hu Yoshida is CTO Emeritus at Hitachi Vantara.

Hu Yoshida

Hu Yoshida

Hu Yoshida spent 24 years at Hitachi Vantara helping define technical direction and enabling customers to address their digital transformation needs. He is widely known in the industry and was instrumental in evangelizing Hitachi's unique approach to storage virtualization.