New system could reduce data-transmission delays across server farms by 99.6 percent
Big websites usually maintain their own "data centers," banks of tens or even hundreds of thousands of servers, all passing data back and forth to field users' requests. Like any big, decentralized network, data centers are prone to congestion:
Packets of data arriving at the same router at the same time are put in a queue, and if the queues get too long, packets can be delayed.
At the annual conference of the ACM Special Interest Group on Data Communication, in August, MIT researchers will present a new network-management system that, in experiments, reduced the average queue length of routers in a Facebook data center by 99.6 percent — virtually doing away with queues.
When network traffic was heavy, the average latency — the delay between the request for an item of information and its arrival — shrank nearly as much, from 3.56 microseconds to 0.23 microseconds.
Like the Internet, most data centers use decentralized communication protocols: Each node in the network decides, based on its own limited observations, how rapidly to send data and which adjacent node to send it to. Decentralized protocols have the advantage of an ability to handle communication over large networks with little administrative oversight.
The MIT system, dubbed Fastpass, instead relies on a central server called an "arbiter" to decide which nodes in the network may send data to which others during which periods of time. "It's not obvious that this is a good idea," says Hari Balakrishnan, the Fujitsu Professor in Electrical Engineering and Computer Science and one of the paper's coauthors.
With Fastpass, a node that wishes to transmit data first issues a request to the arbiter and receives a routing assignment in return. "If you have to pay these maybe 40 microseconds to go to the arbiter, can you really gain much from the whole scheme?" says Jonathan Perry, a graduate student in electrical engineering and computer science (EECS) and another of the paper's authors. "Surprisingly, you can."
Division of labor
Balakrishnan and Perry are joined on the paper by Amy Ousterhout, another graduate student in EECS; Devavrat Shah, the Jamieson Associate Professor of Electrical Engineering and Computer Science; and Hans Fugal of Facebook.
The researchers' experiments indicate that an arbiter with eight cores, or processing units, can keep up with a network transmitting 2.2 terabits of data per second. That's the equivalent of a 2,000-server data center with gigabit-per-second connections transmitting at full bore all the time.
"This paper is not intended to show that you can build this in the world's largest data centers today," Balakrishnan says. "But the question as to whether a more scalable centralized system can be built, we think the answer is yes."
Moreover, "the fact that it's two terabits per second on an eight-core machine is remarkable," Balakrishnan says. "That could have been 200 gigabits per second without the cleverness of the engineering."
The key to Fastpass's efficiency is a technique for splitting up the task of assigning transmission times so that it can be performed in parallel on separate cores. The problem, Balakrishnan says, is one of matching source and destination servers for each time slot.
"If you were asked to parallelize the problem of constructing these matchings," he says, "you would normally try to divide the source-destination pairs into different groups and put this group on one core, this group on another core, and come up with these iterative rounds. This system doesn't do any of that."
Instead, Fastpass assigns each core its own time slot, and the core with the first slot scrolls through the complete list of pending transmission requests. Each time it comes across a pair of servers, neither of which has received an assignment, it schedules them for its slot. All other requests involving either the source or the destination are simply passed on to the next core, which repeats the process with the next time slot. Each core thus receives a slightly attenuated version of the list the previous core analyzed.
Today, to avoid latencies in their networks, most data center operators simply sink more money into them. Fastpass "would reduce the administrative cost and equipment costs and pain and suffering to provide good service to the users," Balakrishnan says. "That allows you to satisfy many more users with the money you would have spent otherwise."
Networks are typically evaluated according to two measures: latency, or the time a single packet of data takes to traverse the network, and throughput, or the total amount of data that can pass through the network in a given interval.
Written by Larry Hardesty, MIT News Office
Andrew Carleen | Eurek Alert!
Information integration and artificial intelligence for better diagnosis and therapy decisions
24.05.2017 | Fraunhofer MEVIS - Institut für Bildgestützte Medizin
World's thinnest hologram paves path to new 3-D world
18.05.2017 | RMIT University
Physicists from the University of Würzburg are capable of generating identical looking single light particles at the push of a button. Two new studies now demonstrate the potential this method holds.
The quantum computer has fuelled the imagination of scientists for decades: It is based on fundamentally different phenomena than a conventional computer....
An international team of physicists has monitored the scattering behaviour of electrons in a non-conducting material in real-time. Their insights could be beneficial for radiotherapy.
We can refer to electrons in non-conducting materials as ‘sluggish’. Typically, they remain fixed in a location, deep inside an atomic composite. It is hence...
Two-dimensional magnetic structures are regarded as a promising material for new types of data storage, since the magnetic properties of individual molecular building blocks can be investigated and modified. For the first time, researchers have now produced a wafer-thin ferrimagnet, in which molecules with different magnetic centers arrange themselves on a gold surface to form a checkerboard pattern. Scientists at the Swiss Nanoscience Institute at the University of Basel and the Paul Scherrer Institute published their findings in the journal Nature Communications.
Ferrimagnets are composed of two centers which are magnetized at different strengths and point in opposing directions. Two-dimensional, quasi-flat ferrimagnets...
An Australian-Chinese research team has created the world's thinnest hologram, paving the way towards the integration of 3D holography into everyday...
In the race to produce a quantum computer, a number of projects are seeking a way to create quantum bits -- or qubits -- that are stable, meaning they are not much affected by changes in their environment. This normally needs highly nonlinear non-dissipative elements capable of functioning at very low temperatures.
In pursuit of this goal, researchers at EPFL's Laboratory of Photonics and Quantum Measurements LPQM (STI/SB), have investigated a nonlinear graphene-based...
24.05.2017 | Event News
23.05.2017 | Event News
22.05.2017 | Event News
24.05.2017 | Physics and Astronomy
24.05.2017 | Physics and Astronomy
24.05.2017 | Event News