A5012092587- Homotopy and Cohomology in Algebraic Topology
- Algebraic structures and combinatorial models
- Algebraic Geometry and Number Theory
- Cloud Computing and Resource Management
- Quantum Computing Algorithms and Architecture
- Parallel Computing and Optimization Techniques
- Quantum Information and Cryptography
- Advanced Algebra and Geometry
- Advanced Topics in Algebra
- Advanced Data Storage Technologies
- Integrated Circuits and Semiconductor Failure Analysis
- VLSI and Analog Circuit Testing
- Fuel Cells and Related Materials
- Caching and Content Delivery
- Electrocatalysts for Energy Conversion
- Low-power high-performance VLSI design
- Heat Transfer and Optimization
- Electrostatic Discharge in Electronics
- Blind Source Separation Techniques
- Green IT and Sustainability
- Network Packet Processing and Optimization
- Interconnection Networks and Systems
- Radiation Effects in Electronics
- Brain Tumor Detection and Classification
- Thermal Analysis in Power Transmission
IBM (United States)
2022
Microsoft (United States)
2014-2021
Rigetti Computing (United States)
2020
Harvard University
2020
Emulate (United States)
2019
Harvard University Press
2018
Illinois Mathematics and Science Academy
2017
Rose–Hulman Institute of Technology
2017
Microsoft Research (United Kingdom)
2013-2016
Microsoft (Finland)
2016
Datacenter workloads demand high computational capabilities, flexibility, power efficiency, and low cost. It is challenging to improve all of these factors simultaneously. To advance datacenter capabilities beyond what commodity server designs can provide, we have designed built a composable, reconfigurablefabric accelerate portions large-scale software services. Each instantiation the fabric consists 6x8 2-D torus high-end Stratix V FPGAs embedded into half-rack 48 machines. One FPGA placed...
Datacenter workloads demand high computational capabilities, flexibility, power efficiency, and low cost. It is challenging to improve all of these factors simultaneously. To advance datacenter capabilities beyond what commodity server designs can provide, we have designed built a composable, reconfigurable fabric accelerate portions large-scale software services. Each instantiation the consists 6×8 2-D torus high-end Stratix V FPGAs embedded into half-rack 48 machines. One FPGA placed each...
Machine learning techniques have led to broad adoption of a statistical model computing. The distributions natively available on quantum processors are superset those classically. Harnessing this attribute has the potential accelerate or otherwise improve machine relative purely classical performance. A key challenge toward that goal is hybridize computing resources and traditional with emerging capabilities general purpose processors. Here, we demonstrate such hybridization by training...
Quantum error correction with erasure qubits promises significant advantages over standard due to favorable thresholds for errors. To realize this advantage in practice requires a qubit which nearly all errors are such errors, and the ability check without dephasing qubit. We demonstrate that “dual-rail qubit” consisting of pair resonantly coupled transmons can form highly coherent qubit, where transmon <a:math xmlns:a="http://www.w3.org/1998/Math/MathML"...
To advance datacenter capabilities beyond what commodity server designs can provide, the authors designed and built a composable, reconfigurable fabric to accelerate large-scale software services. Each instantiation of consists 6 x 8 2D torus high-end field-programmable gate arrays (FPGAs) embedded into half-rack 48 servers. The deployed in bed 1,632 servers FPGAs production successfully used it ranking portion Bing Web search engine by nearly factor two.
In order to support near-term applications of quantum computing, a new compute paradigm has emerged--the quantum-classical cloud--in which computers (QPUs) work in tandem with classical (CPUs) via shared cloud infrastructure. this work, we enumerate the architectural requirements platform, and present framework for benchmarking its runtime performance. addition, walk through two platform-level enhancements, parametric compilation active qubit reset, that specifically optimize architecture...
Abstract Running quantum algorithms protected by error correction requires a real time, classical decoder. To prevent the accumulation of backlog, this decoder must process syndromes from device at faster rate than they are generated. Most prior work on time decoding has focused an isolated logical qubit encoded in surface code. However, for code, programs utility will require multi-qubit interactions performed via lattice surgery. A large merged patch can arise during surgery — possibly as...
When Sean James, who works on data-center technology for Microsoft, suggested that the company put server farms entirely underwater, his colleagues were a bit dubious. But had earlier served board submarine U.S. Navy, submerging whole data centers beneath waves made perfect sense. . This tactic, he argued, would not only limit cost of cooling machines-an enormous expense many operators-but it could also reduce construction costs, make easier to power these facilities with renewable energy,...
For a native gate set which includes all single-qubit gates, we apply results from symplectic geometry to analyze the spaces of two-qubit programs accessible within fixed number gates. These techniques yield an explicit description this subspace as convex polytope, presented by family linear inequalities themselves via finite calculation. We completely describe in variety familiar example cases, and consequence highlight certain member ``<mml:math...
A significant fraction of data stored in cloud storage is rarely accessed. This referred to as cold data; cost-effective for has become a challenge providers. Pelican rack-scale hard-disk based unit designed the basic building block exabyte scale data. In Pelican, server, power, cooling and interconnect bandwidth resources are provisioned by design support workloads; this right-provisioning significantly reduces Pelican's total cost ownership compared traditional disk-based storage.Resource...
Datacenter workloads demand high computational capabilities, flexibility, power efficiency, and low cost. It is challenging to improve all of these factors simultaneously. To advance datacenter capabilities beyond what commodity server designs can provide, we designed built a composable, reconfigurable hardware fabric based on field programmable gate arrays (FPGA). Each in the contains one FPGA, FPGAs within 48-server rack are interconnected over low-latency, high-bandwidth network. We...
Quilc is an open-source, optimizing compiler for gate-based quantum programs written in Quil or QASM, two popular programming languages. The was designed with attention toward NISQ-era computers, specifically recognizing that each gate has a non-negligible and often irrecoverable cost program's successful execution. Quilc's primary goal to make authoring software simpler exercise by making architectural details less burdensome the author. Using allows one write faster while usually not...
We describe an optimal procedure, as well its efficient software implementation, for exact and approximate synthesis of two-qubit unitary operations into any prescribed discrete family XX-type interactions local gates. This arises from the analysis manipulation certain polyhedral subsets space canonical Using this, we analyze which small sets cause greatest improvement in expected infidelity under experimentally-motivated error models. For circuit Haar-randomly selected operations, find...
Customers who operate large data centers are looking for new ways to reduce their energy consumption and improve the ability support higher compute density. An approach liquid cooling that uses warm water instead of chilled cool system components eliminate or greatly need chillers in center is described this paper. This change would have dual effect reducing infrastructure by as much 50% while increasing supported power densities (relative air cooled solutions). In paper a description hybrid...
Fuel cells are a promising power source for future data centers, offering high energy efficiency, low greenhouse gas emissions, and reliability. However, due to mechanical limitations related fuel delivery, slow adjust sudden increases in center demands, which can result temporary shortfalls. To mitigate the impact of shortfalls, prior work has proposed either perform capping by throttling servers, or leverage storage devices (ESDs) that temporarily provide enough make up shortfall while...
To improve the reliability and energy efficiency of datacenters, as well to reduce infrastructure costs environmental impacts, we demonstrated evaluated use a 10 kW Proton Exchange Membrane Fuel Cell (PEMFC) stack system for powering servers in data center. In this study, designed, tested PEMFC Distributed Generation (DG) prime mover that has high both steady state dynamic operations. The 10kW was designed power server rack eliminate distribution datacenter. electrical properties such...
We consider the use of fuel cells for powering data centers, based on benefits in reliability, capital and operational costs, reduced environmental emissions. Using effectively centers introduces several challenges we highlight key research questions designing a cell center power distribution system. analyze specific configuration design space to quantify cost large scale center, most mature commonly deployed technology, achieving over 20% reduction costs using conservative projections.
We consider the use of fuel cells for powering data centers, based on benefits in reliability, capital and operational costs, reduced environmental emissions. Using effectively centers introduces several challenges we highlight key research questions designing a cell center power distribution system. analyze specific configuration design space to quantify cost large scale center, most mature commonly deployed technology, achieving over 20% reduction costs using conservative projections.
In this study, the National Fuel Cell Research Center is working with Microsoft to investigate a new data center concept that includes detailed integration of fuel cell technology into server racks. The could eliminate need for backup power and potentially introduce significant emissions reductions energy savings while enhancing availability. was experimentally verified using hydrogen-fueled 10 kW proton exchange membrane stack system as distributed direct current source, eliminating...