改造数据中心以适应量子计算机.pdf

1、Kristine Rezai,Ph.D.Technical Pre-Sales EngineerIQM Quantum ComputingRetrofitting Data Centers for Quantum ComputersIntractable ProblemsThere is a class of commercial problems that can never be solved with classical computers.Fundamental LimitsThere are not enough atoms in the universe to build a la

2、rge enough memory.Time&Power LimitsComputing the answer would take billions of years with MWs of power consumption.Current approach is not scalable!Why Quantum Computing?Run Time(RT)Problem SizeClassicalBenchmarkQuantum algorithm“Latest findings suggest global computing is more likely responsible fo

3、r between 2.1%and 3.9%of greenhouse gas emissions.“The Energy ChallengeThere is no way we are going to grow computing power as needed.Current approach of increasing computing power is not scalable and already today computing is consuming an increasingly large share of global energy capacity.It is im

4、possible to grow the required computing power sustainably with todays technology1.E+221.E+201.E+181.E+1620102020203020402050New computers trajectories for 1000 ZIPS3456789101 ZIPS1a/bitSource:https:/www.semiconductors.org/wp-content/uploads/2020/10/Decadal-Plan_Interim-Report.pdfZIPS:1021compute ins

5、tructions per secondWhy combine quantum and supercomputing?QuantumSupercomputingHPC+QCNew capabilities for optimization,chemistry+materials simulation+quantum machine learning(QML)Strong at parallelized problems,big dataGeneral purpose computingStrong at high complexity,low data problems(i.e.traveli

6、ng salesperson)Unique superpowers of superposition and entanglement“While quantum computers can accelerate purely classical computing,traditional HPC is simultaneously crucial for getting the most out of quantum acceleration.”Consensus from ADAC16(ORNL,Riken,CSC)https:/arxiv.org/pdf/2508.11765v1Glob