Navops talk at hpc in the cloud meetup 19 march 2019
0 likes92 views
This document discusses how Univa provides software-defined computing infrastructure solutions to help companies modernize workloads and accelerate hybrid cloud migration. It summarizes Univa's products and capabilities, including: - Cluster management and workload orchestration software used by 250+ global customers across industries like energy, government, finance, and more. - Navops Launch, which can deploy and manage hybrid or dedicated HPC clusters across multiple public clouds, and automates all aspects of cloud usage. - Case studies of companies like Wharton School and Mellanox who were able to avoid infrastructure costs and improve resource utilization by bursting workloads to the cloud using Univa's solutions.
Navops talk at hpc in the cloud meetup 19 march 2019
- 1. HPC Migration to the Cloud
- 2. Company Profile • Software-defined computing infrastructure (SDI) solutions to modernize mission- critical workloads and accelerate hybrid cloud migration • Focused on Global 2000 customers across every type of infrastructure and software stack • Thousands of diverse applications, workflows and millions of tasks every day • Offices in Toronto, Chicago and Munich 2 Largest independent provider of cluster management and orchestration software 250+ Global Customers 1,000 Integrated Applications 3,500,000+ Compute Cores 8,000,000+ Jobs Per Day
- 3. Univa market presence by industry 3 Industry leaders use Univa products to optimize performance and manage mission-critical applications Energy Gov’t Financial Life Science Manufacturing Transportation
- 4. Univa product lines 4 Univa provides cluster management and workload orchestration solutions that optimize legacy infrastructure and facilitate hybrid cloud operations Maximize Existing Resources High-performance Computing (HPC) Modernize Cloud Strategy Cloud & Hybrid-Cloud Infrastructure
- 5. What is Navops Launch? Capabilities • Deploy and manage hybrid or dedicated HPC clusters • Fully automated management of cloud VMs • Multi-cloud support: AWS, Azure, GCP, Oracle, OpenStack, VMware Key Features • Policy-based launching and deprovisioning of cloud instances • Automate all aspects of cloud usage • Deep integrations to cloud provider HPC oriented offerings • Integrates well with existing “brownfield” corporate IT tools
- 7. Software Stack - RPMs - Meta Data • Config - Components • UGE Master • UGE Exec • Hardware Profile(s) Navops Launch profiles Hardware Profiles Software Profiles Google Compute Microsoft Azure Amazon EC2 Bare Metal VMware OpenStack Resource Adapters Hardware Stack - Name Format - Install Type - Meta-Data - Resource Adapter(s) Resource Adapter Profiles EC2 1 EC2 2 EC2 3
- 8. Seamless Hybrid Cloud Virtual Machines Bare metal On Prem. Storage Traditional Batch / Machine Learning / Analytic Workloads - Containerized and Non-containerized Cloud Provider API Cloud provider provisions and de-provisions instances Cloud Infrastructure Cloud hosts scale based on workload demand and provisioning policies Data Management Cache, synch, mirror, transfer,.. VPN, VPC Local network extended to cloud-based resources Cloud Storage On-premise infrastructure 8 Policy engine / Cloud automation
- 9. Dedicated Cloud Traditional Batch / Machine Learning / Analytic Workloads - Containerized and Non- containerized Cloud automation engine Cloud Provider API Cloud provider provisions and de-provisions instances Cloud Infrastructure Cloud hosts scale based on workload demand and provisioning policies Cloud Storage Cloud Object Storage
- 10. Policy-based Cloud Automation “With Navops Launch you can easily create custom Automation Applets to dynamically burst and manage your hybrid cloud saving money and driving cloud efficiencies” • Workloads burst based on configurable policies • Scale up, scale down • Move data to the right machine for the right workload • Notify users of specific cloud activity
- 11. CHALLENGE: SOLUTION: CUSTOMER VALUE: Case study: hybrid cloud bursting Wharton was able to avoid substantial infrastructure costs and user training while transparently tripling its core count • Diverse research & analytic workloads • Growing demands by students and faculty • Workload requirements hard to predict • Challenges growing HPC infrastructure • Univa Grid Engine • Navops Launch • Seamless bursting to public cloud GPUs and TensorFlow • Significant cost avoidance on new infrastructure • Improved capacity and strategic flexibility • Hybrid, pay-per-use model for cost efficiency The Wharton School is globally known for intellectual leadership and innovation in business education. 11
- 12. Wharton School of Business 12 Hybrid Cloud Business Drivers Competitive advantage. Get more work done. No waiting. Save Operating Costs. No operating costs when cloud not used. Save Capital Costs. Right size on-premise cluster for optimal utilization not for peak. Save Capital Costs. “Rent” specialized resources like GPUs when required
- 13. CHALLENGE: SOLUTION: CUSTOMER VALUE: Case Study: Hybrid Cloud Bursting Improving EDA license usage by bursting jobs selectively • Critical simulation & verification workloads • Need to defer/reduce CapEx • Unmet demand during peak periods • On-premises EDA licenses underutilized • Univa Grid Engine • Univa License Orchestrator • Univa Unicloud with auto-scaling • Hybrid cloud environment • Avoid resource bottlenecks during tape-out periods • Reduce cost by paying for capacity only when needed • Maximize on-premises EDA license usage by shifting non-licensed work to the cloud, improving productivity By leveraging Navops Launch, Mellanox can extend their local cluster to the cloud, providing increased capacity and maximizing on-prem EDA license usage by workloads to the cloud.
- 14. Western Digital – Million Core Run 14 Gaining Competitive Advantage through Extreme Scale • Reduced time to results from 20 days to 8 hours for critical simulation workloads • Ran 2.3 million jobs on 44,000 AWS spot instances (1+ million cores) with 25 different instance types • Spot instances save significant costs but come and go which created new technical challenges. These were conquered.
- 15. THANK YOU