Patrick M. Oconnor

About Patrick M. Oconnor

An AI data center is a specialized facility designed to support the unique requirements of AI workloads. These data centers house advanced IT infrastructure necessary for training, deploying, and delivering AI applications and services. They differ significantly from traditional data centers in terms of computing power, storage, networking, and cooling capabilities.AI data centers require high-performance computing (HPC) capabilities, often utilizing AI accelerators such as GPUs, NPUs, and TPUs. These accelerators speed up machine learning and deep learning models, making them essential for AI operations.The storage architecture in AI data centers must handle vast amounts of data with high-speed memory. Solid-state drives (SSDs), particularly NVMe SSDs, are critical for their speed and capacity. High-bandwidth memory (HBM) is also used to enable efficient data transfer with lower power consumption.The storage architecture in AI data centers must handle vast amounts of data with high-speed memory. Solid-state drives (SSDs), particularly NVMe SSDs, are critical for their speed and capacity. High-bandwidth memory (HBM) is also used to enable efficient data transfer with lower power consumptionNetworking in AI data centers must support high-bandwidth requirements with low latency. Virtualized network services and software-defined overlay networks optimize compute, storage, and networking for each application and workload. Security and data privacy are also paramount, given the large volumes of data used to train AI modelsAI data centers consume massive amounts of electrical power and require advanced cooling systems to prevent outages and downtime. High-density setups, liquid cooling, and hot/cold aisle containment are some methods used to manage these requirements. Sustainability is also a consideration, with some data centers running entirely on renewable energyHyperscale data centers are large facilities with at least servers and square feet of space. They provide extreme scalability and are used by cloud providers like AWS, Microsoft Azure, and Google Cloud Platform for AI, data analytics, and other large-scale workloadsColocation data centers are owned by one company but rented out to others. This setup allows businesses to benefit from hyperscale capabilities without significant investment. Major cloud service providers often lease space from colocation operators and rent it out to other businesses

Experience

1. Edge Regional Data Center IT Operations Management

   Microsoft

   Dec 2021 — Present · US

   Member of Microsoft\'s Global Data Center Operations team.

Education

• The University of Texas at Austin

  B.A. History

Skills

• Facility Security
• Telecommunications
• Hvac
• Diesel Genertors
• Itil
• Load Balancers
• Ip
• Ssae 16 Audits
• Data Center Manager
• IT Service Management
• Dc Power Plants
• Technology Implementation
• Data Center Operations
• Security
• IT Service Delivery
• IT Management
• Bicsi
• Backups
• Computer Network Operations
• Leadership Mentoring
• Vendor Management
• Service Delivery
• Optical Fiber
• Management
• Switches
• Firewalls
• Network & Voice
• Cabling
• Building Automation
• Ethernet
• Managed Services
• Dns
• Integration
• Ups Systems
• Logical Security
• Contract Negotiation
• Project Management
• Data Center Engineer
• Networking
• Operations Management