Why Nutanix for Enterprise Workloads
2 likes1,635 views
The document discusses the capabilities and advantages of Nutanix's hyperconverged infrastructure for enterprise workloads, emphasizing the importance of data locality and features like metro availability and storage services. It outlines key functionalities such as virtual machine management, disaster recovery, and integration with public cloud services, as well as enhancements like automated failover and network orchestration. Additionally, it covers tools for efficient management of resources and ensuring scalability in dynamic environments.
Why Nutanix for Enterprise Workloads
- 1. Why Nutanix for enterprise workloads Sr Systems Engineer Switzerland
- 2. 22 Nutanix Basics Why Data Locality matter Metro Availability Witness Acropolis Services PRISM Central Micro Segmentation Cross Hypervisor DR Self Service Portal / NuCalm Agenda
- 3. 33 Nutanix Basics Why Data Locality matter Metro Availability Witness Acropolis Services PRISM Central Micro Segmentation Cross Hypervisor DR Self Service Portal / NuCalm Agenda
- 4. 4 Virtualization App App Integrated, scale-out compute and storage Virtualization App App Storage Controller Storage Controller Storage Controller Storage Controller Server Server Storage Controller Storage Controller with built-in virtualization and management Re-Platforming the Enterprise Datacenter
- 5. 5 VM VM VM CVM Hypervisor VM VM VM CVM Hypervisor Tier 1 Workloads (running on all nodes) Nutanix Controller VM (one per node) VM VM VM CVM Hypervisor Distributed Storage Fabric Snapshots Clones Compression Deduplication Locality Tiering Erasure Coding Resilience Node 1 Node 2 Node N X86 X86 X86 Scalable Distributed System Design
- 7. 7 Integrated Compute & Storage VM-centric Storage Backup & DR Services Hyperconverged Infrastructure Native Virtualization Multi-hypervisor Support Hybrid-cloud Services Machine Intelligence Systems and Operations Management File and Block Storage HCI++ Container Support Cloud Automation* Application Mobility Integrated Public Cloud Microsegmentation* Object Storage* Application Ecosystem* Enterprise Cloud Continuous Innovation with the Enterprise Cloud OS
- 8. 88 Nutanix Basics Why Data Locality matter Metro Availability Witness Acropolis Services PRISM Central Micro Segmentation Cross Hypervisor DR Self Service Portal / NuCalm Agenda
- 9. 9 NVMe Replaces SATA and SAS interfaces with a PCIe-based standard Fabric topologies replace networks 3D XPoint (Optane) New nonvolatile storage memory from Intel and Micron Much greater performance than flash (1000X) Trends: Increased Performance
- 10. 10 0 10000 20000 30000 40000 50000 60000 ThroughputinMB/S Flash Network SSD 10G NVMe 40G 100G 3DXPoint When Applications predominantly access data locally, NW bandwidth demands are lowered Why Care About Data Locality
- 11. 1111 Importance of Auto-Tiering and Data Locality
- 12. 1212 Nutanix Basics Why Data Locality matter Metro Availability Witness Acropolis Services PRISM Central Micro Segmentation Cross Hypervisor DR Self Service Portal / NuCalm Agenda
- 13. 1313 13 • Automates failover decisions for Metro Availability deployments • Zero RPO with near-Zero RTO • Automatic failure handling with manual recovery • Witness handles common failure scenarios: • Site failure • Avoids split brain scenarios Metro Availabilty Witness
- 14. 1414 Site 1 Site 2 Site 3 Witness • Stand-alone Witness VM • Deploy via OVF on 3rd site or separate fault domains • Does not need to run on Nutanix hardware • One shared Witness VM per cluster. • Scalability up to 50 Protection Domains per Witness • Connections to Witness VM need to be independent from inter-site connection to ensure separate failure domain location for Witness VM • Witness connection < 200ms latency RTT • Inter-site connection < 5ms latency RTT • Nutanix Witness VM requires: • 2 vCPUs • 4 GB of Memory • 25 GB of storage Latency < 5ms RTT Latency < 200ms RTT Latency < 200ms RTT Three independent network connections Metro Availabilty Witness - Details
- 15. 1515 Nutanix Basics Why Data Locality matter Metro Availability Witness Acropolis Services PRISM Central Micro Segmentation Cross Hypervisor DR Self Service Portal / NuCalm Agenda
- 16. 16 AHV provides what is “needed” and when combined with Nutanix AMF and DSF, is a comprehensive virtualization platform that is much more than a “hypervisor” • Factory Secured / Self-healing compliance • Cross Hypervisor DR / Migration • Self Service Provisioning Portal • OpenStack Integration • Hybrid Cloud Support (AWS, Azure) • Scalable / Multi-site Management • Operational Insights with Predictive Analytics • File & Block Services Hypervisor vMotion VM Migration Hot Add dVS EVC Hot Add CPU Compatibility Open vSwitch vVols VM Centric DSF CBT Replication CBT Replication Management Host HA Host HA VM HA ADS / VM HA DRS Hot Spot Avoidance Host Profiles Foundation VUM 1-Click Upgrades Storage AHV where we are
- 20. 20 Cloud Connect Local VM Snapshots (On-cluster) Single-Node Target (On-Prem) (ROBO only) Backup/DR over WAN to Main DC ROBO/ SMB AWS Overview Backup/Replication for ROBO/SMB Leverage Nutanix Native Snapshots Up to 40 TB Raw Capacity Cannot be used as a generic backup target Benefits Cost effective alternative to traditional options Easy & Fast Self Service Restores Simple & unified Prism Management Efficient storage utilization with support for Compression and deduplication Single Node Replication Target
- 21. 2121 Nutanix Basics Why Data Locality matter Metro Availability Witness Acropolis Services PRISM Central Micro Segmentation Cross Hypervisor DR Self Service Portal / NuCalm Agenda
- 22. 22 How to efficiently manage VMs and clusters at scale Nutanix Prism gives IT a centralized view Centralized Management with Prism
- 23. 23 • Storage Management • Cluster Management • VM Management • Virtual Network Management One-click Infrastructure Management • Capacity Behavior Trends • Capacity Optimization Advisor One-click Planning • Alert Analysis and Recommendations • Customizable Dashboards and Charts One-click Operational Insights Datacenter Management with Prism
- 24. 2424 • Fundamental to mixed hypervisor support • Single management GUI for AHV and ESXi • Use Prism for common use cases • VM Create • VM Update • VM Delete • VM Power On/Off Operations • Launch console • Clone • Does not displace vCenter VM Management for ESXi
- 25. 25 What is it? • Detailed capacity trends • Intelligent recommendations • Powered by patent pending X-Fit algorithm Benefits: • Enables pay as you grow capacity expansion • Eliminates guesswork Actionable Capacity Forecasting
- 26. 26 3 4 vCPU, 4GB, CPU intensive 2 2 vCPU, 8GB, IO intensive 4 vCPU, 8GB, memory intensive VDI XenApp Exchange SQL Server Splunk • What-if for a pre-defined workload type • What-if for a group of VMs with different configurations and behaviors • What-if for an expected demand change of existing running workloads. • What-if for increasing the runway • What-if for new Infrastructure What customers get Ensure the capacity expansion to meet the future workload growth. 50% More of current consumption Machine-learned behavior What-if analysis Recommendation What-if Analysis
- 27. 2727 Nutanix Basics Why Data Locality matter Metro Availability Witness Acropolis Services PRISM Central Micro Segmentation Cross Hypervisor DR Self Service Portal / NuCalm Agenda
- 30. 31 • View network design up to TOR • See real-time connection health • Monitor network link performance • Instantly identify VLAN misconfigurations • Filter by applications, hosts, users, VMs and VLANs Network Visualization
- 33. 34 Network OrchestrationTOR TOR Network Tap TOR TOR Load Balancer • Instant Network Provisioning • Automated L4 to L7 Integration • Service Insertion • Network Taps Network Orchestration
- 35. 3737 Nutanix Basics Why Data Locality matter Metro Availability Witness Acropolis Services PRISM Central Micro Segmentation Cross Hypervisor DR Self Service Portal / NuCalm Agenda
- 36. 3838 Main Data Center DR Site Cross Hypervisor DR
- 37. 3939 Use cases for Cross Hypervisor DR: • Easy migration to AHV. • Cost reduction for DR sites • ROBO replication into central DC • SPs offering DRaaS using AHV in their DC No competing solution. Why? • Complex replication and conversion • High learning curve with various migration tools Cross Hypervisor DR: Usecases
- 38. 4040 ESX AHV 1-Click Hypervisor Conversion
- 39. 4141 • 1-click conversion from ESX cluster to AHV • - • Easier onboarding of workloads to AHV. - Start with ESXi - Deploy VMs - Convert to AHV • Ease of evaluation of AHV 1 click conversion of a test/dev cluster to AHV. In Place Hypervisor Conversion: Why?
- 40. 4343 Nutanix Basics Why Data Locality matter Metro Availability Witness Acropolis Services PRISM Central Micro Segmentation Cross Hypervisor DR Self Service Portal / NuCalm Agenda
- 41. 4444 Nutanix Admin VLAN backed Virtual networks Users Images Projects Project 1 Quota 1 Project 2 Quota 2 Self Service Portal Prism Self Service Portal
- 42. 45 Deploy Applications (VM CRUD) Monitor Applications Monitor Resource Usage • Role based access with AD integration • VM User role • VM View role • Catalog items • VM Template • vDisk • Images from Docker hub • App templates will be available soon after • Monitor deployed applications Tenant Capabilities
- 43. 47 Day 2 and Beyond: App LifeCycle Management Patches & Upgrades Cost Showback Monitoring The NuCalm Experience
- 44. 4848 Q & A