Industry executives and experts share their predictions for 2023. Read them in this 15th annual VMblog.com series exclusive.
The Rise of Computational Storage, Data Explodes, and How Data Center Sustainability is Transforming Storage
By Hao Zhong,
co-founder, and CEO of ScaleFlux
Prediction: The Time for Computational
Storage to go Mainstream is Upon us
Computational storage (CS) has been
discussed for years, but it has yet to reach high levels of mainstream
adoption. In the coming year, however, most large enterprises will embrace this
technology with open arms.
As enterprises generate more data,
they're increasingly faced with the reality that "scale-out" is not a
long-term sustainable strategy. Data center space is at a premium, and every
node added triggers cascading costs and architecture challenges. The fixed size
of today's storage devices is already creating problems. It has become
increasingly complicated and expensive to evolve workloads: As they grow or
change, the only option is often to add additional drives.
As we see it, commodity flash storage is
not up to the task in the modern data center. Thankfully, we will have an
alternative to commodity storage that alleviates many of these issues by
offloading the server's central processing unit (CPU) so that enterprises can
get more functionality from their servers, including better endurance and
performance, higher capacity, and cost savings. As a result, we expect to see
the adoption of easy-to-use computational storage grow significantly in 2023,
as now there is a real choice between commodity solid-state drives (SSDs) and
SSDs with computational storage technology packaged as native NVMe-compatible
drives.
Prediction: Data Growth Fuels a Rise
in Edge Computing
All industries are experiencing surging data growth, a
ubiquitous problem that has become both a cliché and a force of nature. That
said, adapting to the complexities of massive data growth introduces many
issues into a data center's architecture, forcing companies to adopt an
entirely different approach to data storage - edge computing. Edge computing
offers a significant reduction in data traffic over the backbone network for a
better quality of service (QoS), both natural effects of moving computation and
storage closer to data sources and end users. As a result, global Industry
analysts predict the Edge Computing Market will reach $21.6 Billion by 2027
(27% higher than 2020).
As companies continue to collect large sums of data,
they've simultaneously lost quick accessibility to good data, a problem that
isn't going away soon. Data creation is projected to reach a staggering 180
zettabytes by 2025 -- 118.8 zettabytes more than in 2020. Edge computing
processes data close to the original source, allowing users to access data
easier, significantly increasing customer experiences and business efficiency.
Some industries, such as 5G, IoT, and autonomous vehicles, have already adopted
edge computing and are continuing to reap the benefits.
The success of edge computing as we
step into 2023 will largely depend on its cost-effectiveness. As semiconductor
technology scaling reaches its limit, heterogeneous and domain-specific
computing becomes critical in maintaining the cost-effectiveness of IT
infrastructure. Another key component of the future heterogeneous computing
paradigms is computational storage, which will serve as an essential building
block for cost-effective edge computing infrastructure. Together, edge
computing and the cloud will complement each other, forming the foundation of a
future pervasive IT infrastructure. In the coming years, we can expect the
continued growth of the cloud in the form of centralized data centers and the
rapid expansion of edge computing.
Prediction: Emerging sustainability
requirements in the data center will transform the storage industry
As enterprises evaluate the total
costs of their infrastructure, there will be an increased focus on improving
sustainability. Newer and more efficient
leased data centers will begin to have the edge over aging, enterprise-owned
facilities. Along with more traditional criteria such as cost and performance,
we will see more enterprises including environmental footprint calculations in
their decision-making for infrastructure strategy.
Storage is one of the obvious domains
to examine as these requirements are implemented. Significant improvements in the lifespan of SSDs
will reduce waste and carbon footprint, especially in write-intensive
environments. Additional savings in
power can be had with better-optimized devices and the reduction of unnecessary
processing on the CPU. Greener technology
that can also deliver on the performance and cost demands in the modern data
center will grab the attention of enterprises of every size and market.
##
ABOUT THE AUTHOR
Dr. Hao Zhong is a passionate technology innovator and entrepreneur that has been contributing numerous cutting edge technology and product in data storage and computing industry. Since Oct 2014, Hao, as the CEO, co-founded ScaleFlux that is leading the computational storage subsystem development in the industry.
Prior to ScaleFlux, Hao was a Sr. Director at Fusion-io where he led the flash memory technology team and enabled LDPC technology in industry leading PCIe SSDs. Prior to Fusion-io, Hao was with SandForce as an engineering director working on SSD controller chips. Prior to SandForce, Hao was an architect at LSI working on hard disk drive read channel product and the key contributor to the industry first 40nm LDPC read channel chip in 2008. Hao received his Ph.D degree from Rensselaer Polytechnic Institute.