The Manager, High Performance Computing will be responsible for the high performance computing and data storage environment for NYGC. The primary responsibilities are to maintain and support a batch computing cluster and a data storage environment. This individual will be the first member of what will expand into a full research computing team and will be responsible for a considerable variety of tasks as NYGC moves from initial operations in early 2012 into full production by the end of the year.
To apply for this position please email a resume and cover letter to [email protected].
Job Responsibilities:
-Configure and support a heterogeneous batch compute cluster of approximately 1,000 CPU cores;
-Configure and support scientific data storage totaling approximately 1PB;
-Configure and support network devices required by the cluster and storage;
-Create written documentation for operational and usage procedures;
-Install and configure genomics and bioinformatics applications;
-Work closely with bioinformatics staff to ensure that the system is both useable and useful;
-Develop and maintain an initial cloud computing system and ensure HPC capabilities are robust throughout the transition from a fully cloud-based system to a system combining on-site computing, off-site data storage, and burst cloud capacity;
-Perform regular benchmarking and performance measurement;
-Participate in hardware and software lifecycle planning, including recommending hardware and software purchases; work with NYGC leadership team to make decisions regarding vendor selection and management;
-Collaborate with other internal and external information technology consultants and advisors.
Position Requirements:
-At least 5 years experience in high performance and cluster computing for scientific applications;
-At least 5 years experience in Unix / Linux system administration;
-Knowledgeable about batch queuing environments including Sun Grid Engine and LSF;
-Excellent Unix / Linux scripting skills for data motion and process automation;
-A high tolerance for change will be beneficial as NYGC is in start-up mode and the environment will morph rapidly;
-Willingness and ability to make recommendations and participate in decision-making;
-Excellent written and verbal communication skills required.
To apply for this position please email a resume and cover letter to [email protected].
Job Responsibilities:
-Configure and support a heterogeneous batch compute cluster of approximately 1,000 CPU cores;
-Configure and support scientific data storage totaling approximately 1PB;
-Configure and support network devices required by the cluster and storage;
-Create written documentation for operational and usage procedures;
-Install and configure genomics and bioinformatics applications;
-Work closely with bioinformatics staff to ensure that the system is both useable and useful;
-Develop and maintain an initial cloud computing system and ensure HPC capabilities are robust throughout the transition from a fully cloud-based system to a system combining on-site computing, off-site data storage, and burst cloud capacity;
-Perform regular benchmarking and performance measurement;
-Participate in hardware and software lifecycle planning, including recommending hardware and software purchases; work with NYGC leadership team to make decisions regarding vendor selection and management;
-Collaborate with other internal and external information technology consultants and advisors.
Position Requirements:
-At least 5 years experience in high performance and cluster computing for scientific applications;
-At least 5 years experience in Unix / Linux system administration;
-Knowledgeable about batch queuing environments including Sun Grid Engine and LSF;
-Excellent Unix / Linux scripting skills for data motion and process automation;
-A high tolerance for change will be beneficial as NYGC is in start-up mode and the environment will morph rapidly;
-Willingness and ability to make recommendations and participate in decision-making;
-Excellent written and verbal communication skills required.