ray (FPGA) 13, 21 and GPU,h with storage media) and flash and other emerging new non-volatile memories (such as 3D XPoint, ReRAM, STT-RAM, and PCM) that provide persistent storage at DRAM latencies to deliver high-performance gains. This approach would present the greatest flexibility to take advantage of advances in the underlying storage device to optimize performance for multiple cloud applications. In the near future, the software-hardware innovation inside the SSD can proceed much like the PC, networking hardware, and h https://bit.ly/2L8LfM4

GPU ecosystems have in the past. This is an opportunity to rethink datacenter architecture with efficient use of heterogeneous, energy-efficient hardware, which is the way forward for higher performance at lower power.

Value Propositions

Here, we summarize three value propositions that demonstrate future directions in programmable storage (see Figure 4):

Agile, flexible storage interface (see Figure 4a). Full programmability will allow the storage interface and feature set to evolve at cloud speed, without having to persuade standardization bodies to bless them or persuade device manufacturers to implement them in the next-generation hardware roadmap, both usually involving years of delay. A richer, customizable storage interface will allow application developers to stay focused on their application, without having to work around storage constraints, quirks, or peculiarities, thus improving developer productivity.

As an example of the need for such an interface, consider how stream writes are handled in the SSD today. Because the SSD cannot differentiate between incoming data from multiple streams, it could pack data from different streams onto the same flash erase block, the smallest unit that can be erased from flash at once. When a portion of the stream data is deleted, it leaves blocks with holes of invalid data. To reclaim these blocks, the garbage-collection activity inside the SSD must copy around the valid data, slowing the device and increasing write amplification, thus reducing device lifetime.

If application developers had con-
trol over the software inside the SSD,
they could handle streams much more
efficiently. For instance, incoming
writes could be tagged with stream
IDs and the device could use this in-
formation to fill a block with data
from the same stream. When data
from that stream is deleted, the entire
data block could be reclaimed with-
out copying around data. Such stream
awareness has been shown to double
device lifetime, significantly increas-
ing read performance. 14 In Micro-
soft, this need of supporting multiple
streams in the SSD was identified in
2014, but NVMe incorporated the fea-
that splits the monolithic components
of an SSD into two different modules—
one standardized part dealing with
storage media and a software interface
to handle application-specific tasks
(such as garbage collection and wear
leveling). In this way, SSD suppliers can
build simpler products for datacen-
ters and deliver them to market more
quickly while per-application tuning is
possible by datacenter operators.

The component-based ecosystem also opens up entirely new opportunities for integrating powerful heterogeneous programming elements (such as field-programmable gate ar-

Figure 4. Programmable SSD value proposition.

Host Server

Programmable SSD (c)

Trusted domain for secure computation
(cleartext not allowed to egress the SSD boundary.)

Host

Server

Programmable SSD (b)

Moving compute inside SSD to leverage
low latency, high bandwidth, and access to data
Server

Programmable SSD

DRAM
Processor +
HW Offload

(a)

Agile, flexible storage interface
leveraging programmability within SSD
DRAM
Processor +
HW Offload
DRAM
Processor +
HW Offload