Tackling the huge amounts of data generated by the Internet
of Things (IoT) is leading to two interesting new hardware and software architectures alongside the current focus on machine learning and artificial intelligence.
HP Enterprise has developed a new technique for handling
large amounts of information in the data centre. The Memory-Driven architecture
makes all the data easily available, and HPE has developed a system with 160TBytes
of such storage.
A key feature of the new architecture is direct photonic
interconnections between each block of memory. The direct fibre connections
from every memory block to every other block means the processor can access
data anywhere in memory at effectively the same speed, which allows the system
to perform calculations far more rapidly. This also allows specialised
processors such as AI accelerators to access the same shared pool of data, such as information
from the Internet of Things.
“We’re all intrigued by the potential of data to change our
lives, but just as we’re ready to take advantage of it, the technologies that got
us this far are petering out,” said Sharad Singhal, director of software and
applications for HPE’s Machine project.
The best way to handle such data, says Singhal, is to put
the whole graph in memory at once, and make all physical blocks of memory
equally quick to access. Memory-Driven Computing makes that much more feasible.
HPE has now launched a system with 160Tbytes of this shared memory, allowing much more flexible and efficient processing of large sets of data.
Open source software
Researchers at the University of Michigan have developed open
source software that provides and efficient way to share server memory in order
to speed up performance of existing hardware in data centres to address this
same problem.
The Infiniswap software boosts Remote Direct Memory Access
network performance by 47% in a cluster without having to change the hardwares.
"Infiniswap is the first system to scalably implement
cluster-wide 'memory disaggregation,' whereby the memory of all the servers in
a computing cluster is transparently exposed as a single memory pool to all the
applications in the cluster," said Infiniswap project leader Mosharaf
Chowdhury, U-M assistant professor of computer science and engineering.
"Memory disaggregation is considered a crown jewel in
large scale computing because of memory scarcity in modern clusters."
The software lets servers instantly borrow memory from other
servers in the cluster when they run out, instead of writing to slower storage
media such as disks. To avoid the memory bottleneck, the Michigan team designed
a fully decentralized structure. With no centralized entity keeping track of
the memory status of all the servers, it doesn't matter how large the computer
cluster is. Additionally, Infiniswap does not require designing any new
hardware or making modifications to existing applications. The research team tested Infiniswap on a 32-machine RDMA
cluster with workloads from data-intensive applications that ranged from
in-memory databases such as VoltDB and Memcached to popular big data software
Apache Spark, PowerGraph and GraphX.
They found that Infiniswap improves by an order of magnitude
both "throughput"—the number of operations performed per second—and
"tail latency"—the speed of the slowest operation. Throughput rates
improved between 4 and 16 times with Infiniswap, and tail latency by a factor
of 61.
"The idea of borrowing memory over the network if your
disk is slow has been around since the 1990s, but network connections haven't
been fast enough," Chowdhury said. "Now, we have reached the point
where most data centres are deploying low-latency RDMA networks of the type
previously only available in supercomputing environments."
- King Cobra rugged server design provides a “total rack” in only 2U
- IoT drives software defined power into the data centre
- Xilinx expands its 16nm UltraScale+ FPGA roadmap for new data centre technology
- Another startup aims for embedded AI
- NVIDIA pushes artificial intelligence into embedded designs with Jetson TX2
No comments:
Post a Comment