Advanced features and capabilities Platform architecture Ayman - - PowerPoint PPT Presentation

Advanced features and capabilities Platform architecture

| 1

Ayman Hamed Solutions Architect

How We Built It: Web-Scale Engineering

Off-the-shelf x86 servers All intelligence in software Scale-out architecture Self-healing systems Rich automation and analytics

X86 X86 X86

Unlimited scalability

• No limitation on the maximum number of node per cluster
• Start as small as you need and scale as you need
• Non-disruptive cluster expansion

Heterogeneous cluster

A true scale-out architecture with no limitation

6

4 x NX-1065-G4 ( 2 x E5 CPUs, 128G RAM, 1 960G SSD, 2 6TB HDDs ) 1 X NX-1075S-G5 ( 1 x E5 CPU, 256G RAM, 2 x 480G SSDs, 2 x 4TB HDDs ) 2 x NX-8035-G6 ( 2 x Skylake CPUs, 384G RAM, 6 x 1.92TB SSDs All Flash ) 1 x node NX-5055-G6 storage-only 1 x NX-8055-G6 ( 2 Skylake CPUs, 786G RAM, 4 x 1.92TB SSDs, 8 x 6TB HDDs )

Tunable Redundancy

What is it

• Configure different levels of fault tolerance

for different applications dynamically

Points of differentiation

• Software-defined. No prior understanding of

storage required

• RF-3 offers protection against two

simultaneous disk, node and NIC failures

• Switch between RF-2 and RF-3
• Replication is set at the

container/application level

CPU Memory CPU Memory CPU Memory CPU Memory

RF-2 Container RF-3 Container

One platform - Multi-site - Multi-cloud

E N T E R P R I S E C L O U D | C O N F I D E N T I A L | 9

Main Site DR Site Branch - AHV Branch on ESXi

1-niode

Prism Central Branch AHV

Data Locality

Unlimited tax-free scale and performance!

Storage Storage Hypervisor Hypervisor Controller VM Controller VM

Anatomy of a Write I/O

Node

Guest VM Hypervisor Controller VM Storage

Node Node

Performance and availability

• Data is written locally
• Replicated on other nodes for high availability
• Replicas are spread across cluster for high performance

Intelligent real-time automated Data Tiering

• Hot data – in the SSD tier for fastest performance
• Cold data in the HDD tier for maximum efficiency
• All SSD and HDD resources are available for all VMs

across the cluster

VM Pinning to SSDs or HDDs

The right workload on the right tier

• Keep Data for critical VMs in the SSD tier for fastest

performance

• Enables Mix IOPS-sensitive and regular workloads in

the same cluster

• Consistent and fast storage performance for

applications such as databases

Advanced features and capabilities Data Services

| 14

Ayman Hamed Solutions Architect

Compression (inline and post-process)

Granular resource management/utilization

 Inline and post-process compression  Inline: Data compressed as it’s written  MapReduce: Data compressed after “cold”

data is migrated to lower-performance storage tiers

10100101 10101010 10100101 10101010 10100101 10101010

10100101 10101010 10100101 10101010 10100101 10101010 10100101 10101010

 No impact to normal IO path  Ideal for random batch workloads  Uses Snappy algorithm

Deduplication (inline and post-process)

Granular resource management/utilization

 “Inline fingerprinting with post-

process deduplication

 Distributed across all nodes –

true scale-out

 Global dedup across entire

cluster

 SHA-1 fingerprinting offloaded to

Intel processors for greater efficiency

 100% software-defined  Strong hash allows dedup to

happen based on metadata match

Reduction Up to

Compression Benefits

Real life examples

AFS AFS

Nutanix Files

NFS/SMB Store Files & Folders FILES

Simple & Intelligent

• Deploys In Minutes
• Fully Managed Service
• 1-Click Insights & Recommendations

Scalable

• Pay As you Grow, Pay Only For What Your Store
• PB Scale, Billions Of Files, Large or Small
• Grow & Shrink Capacity & Performance As Needed

Enterprise Class

• Standard Interfaces - NFS & SMB
• Secure, Reliable and Resilient
• Support 1000s of Connections

Enterprise Cloud OS

Nutanix Volumes

What is Nutanix Volumes ?

– Exposes storage through direct block- level iSCSI access – Nutanix Volumes enables high availability, scalability, and performance to non-virtualized Oracle, SQL, and IBM DB2 workloads

20

iSCSI iSCSI

Non-virtualized Servers Nutanix Nodes

Advanced features and capabilities Data Protection Backup, DR and HA

| 21

Ayman Hamed Solutions Architect

A Complete Data Protection solution

Nutanix Local and Remote Snapshots

Overview

 Create unlimited local copies of VM  Allow policy-based snapshot management  Enable app and crash consistent

policies

 Leverage Prism for snapshot management  Support for multiple hypervisors

Benefits

 Self-service file-level restore  VM to LUN mapping eliminated  No performance impact through redirect on

write

 Efficient storage utilization  Unified consumer grade interface

VM Primary Cluster Local VM-Centric Snapshots

DR Cluster Remote VM Snapshots

RTO: Minutes RPO: Minutes

Application Consistent Snapshots

24

Overview

 Application-aware snapshots for Windows and

Linux

 Integrates with VSS on Windows  Pre & Post script hooks available on Windows

and Linux

 Hooks provide even deeper application

integration

Benefits

 Fast recovery on restore – no replays required  Guaranteed application consistent state on disk  Support for any application on Linux and Windows Snapshot

call Pre Script Hook

quiesce app, trunc logs, etc

Post Script Hook VSS* Snapshot Snapshot return

* Windows OS support only

Async Backup and DR Replication scenarios

Cloud Connect Local VM Snapshots (On-cluster) Single-Node Target (On-Prem) (ROBO only) Backup/DR

• ver WAN to Main DC

ROBO/ SMB

AWS/Azure

Replication scenarios:

 Replication to local

containers

 Replication to remote cluster  Replication to a backup

target

 Replication to the Cloud

Benefits

 Data Redundancy at SSD/HDD tier  Cost effective alternative to

Nutanix Self Service Restore (SSR)

Overview

 Recover independent files from VM

Snapshots

 Multi-tenant in nature

Benefits

 Self-service recovery without Admin

intervention

 Eliminates the need to recover an entire

VM

 Easy to set up and manage VM Owner Nutanix Admin

• VM Owner can only see his/her VM

snapshots (multi-tenant)

• Can mount snapshot (drive letter), then

recover files without Admin involvement

• Gives control of VM snapshots to VM
• wner

VM Snapshots

Nutanix Cloud Connect

Overview

 Enable Hybrid Cloud deployment  Backup VMs to Public Cloud on a user defined

schedule

 Leverage Prism for natively integrated

management

 Support for AWS and Azure

Benefits

 Unlimited and efficient cloud-based data storage  WAN optimized with over the wire compression &

dedup

 Flexible policy management with RPO in minutes  Interoperability with Nutanix DP/DR portfolio

RTO: Minutes RPO: Minutes DR Cluster Remote VM Snapshots vdis k

Local VM-Centric Snapshots

Primary Cluster

Disaster Recovery

28

Asynchronous Replication

Overview

 Asynchronously replicate VMs  Policy based replication  Incremental byte-level transfer  Cross-hypervisor replication support  Multi-site Replication Support  Multi-topology Replication support

Benefits

 Support for multiple retention and

scheduling policies

 WAN-optimized to eliminate network

throttle

 Centralized management via Prism  One Click management using Prism

RTO: Minutes RPO: Minutes DR Cluster Remote VM Snapshots Location 2 Location 3

Protection Domain 1 Protection Domain 2 Remote Site Remote Site Remote Site Remote Site Remote Site

vdis k

Local VM-Centric Snapshots

Primary Cluster Location 1

Near Sync Replication

Overview

 Replication optimization that allows 1-minute

RPO for Mission-critical Applications.

 No restrictions on latency or distance  Leverages Nutanix Light-Weight Snapshots (LWS)  Leverages existing Protection Domain-based

workflows

Benefits

 Minimal data loss in case of disaster  Highly granular restore capabilities.  Extremely simple to setup and failover  Comparable RPO to high end arrays  Simple configuration with time based retention

policy

RTO: Minutes RPO: 1 Minute Local Cluster Remote Cluster S0 S0 LWS LWS S1

1 MIN

S1

*Adhere sizing requirements

• Enjoy hypervisor choice while using integrated backup and DR
• All the benefits of Nutanix async remote replication technologies
• Automatic migration of configuration and VM conversion

Value Proposition and Benefits

AHV AHV

Cross-Hypervisor Backup and DR

Nutanix Xi (Leap): Cloud Services

| 32

• Enable Cloud-based DR with a few clicks
• Automatically migrate apps and data to Xi Cloud
• One-click testing and failover
• Flexible subscription plans

Primary Site

Security Network Security Network Live VMs Offline VMs

Nutanix Metro Availability Cluster

Site 1 Site 2 Site 3

Witness

Overview

 Automates failover decisions for Metro

Availability Real-time copies of data across two datacenters

 Provides ‘outside view’ into difference

between a network partition and a remote site failure

 Located in separate failure domain

Benefits

 Continuous availability of applications and

data

 Avoid Split-Brain scenarios to prevent data

corruption

 Operational simplicity with Prism interface  Scalable up to 50 containers/protection

domain

RTO: Near Zero RPO: Zero

Nutanix Data Protection, HA and DR

Advanced features Security

| 35

Husain Ebrahim

• Sr. Systems Engineer

36

Security-First Design

Security Lifecycle

Segmentation Encryption Anti-Virus Anti-Malware Firewall Key Management

Features and Ecosystem Standards and Certifications Automated Validation and Self-healing

Nutanix Data at Rest Encryption Portfolio

Self Encrypting Drives Nutanix Nodes External Key Manager Regular Drives

Old Solution

(still supported)

With AOS 5.5

(Get rid of SEDs) External Key Manager Regular Drives Nutanix Nodes with SW Encryption

With AOS 5.8

(Get rid of both SEDs and EKM) Nutanix Nodes with SW Encryption & Native Key Mgmt.

Data at Rest Encryption

KMS

Choose Native Key management

• r

Nutanix Ready KMS Providers

KMIP

Flow – Visibility, Security, Automation

TCP:8443 TCP:1575 TCP:443

Application Visualization Microsegmentatio n

(East-West Firewall)

• APM
• Load Balancer
• vFirewall

Service Chains Network Functions

Advanced features Automation and Operations Management

| 4

Husain in Ebra rahim im

• Sr. Systems Engine

ineer

Virtualization Management Compute, Storage, Networking, Data Protection Management Operational Analytics

Prism: Full-Stack Management & Analytics

Cluster Health

Firmware OS Hypervisor Software

Non-Disruptive, Dependency-Aware Upgrades

Full Lifecycle Management

| 43

Buy Deploy Manage

Troubleshoot

Upgrade Scale

Scale without Limit

• Add one node at a time
• Automatic redistribution
• 100% linear performance

Easy Deployment

• Quick cluster install
• Auto-discovered nodes
• Simple infra sizing

Simple Management

• Single pane of glass
• No IT expertise required
• Access on any device

Real-time Insights

• Machine Learning engine
• Capacity Planning
• Anomaly detection

One-Click Upgrades

• Nutanix software
• Hypervisor
• Firmware and BIOS

Prism Self-Service

What is it?

• Self-service access to IT resources
• IT sets policies and provides access to

end users such as developers

• Components:

1.

Admin Portal: Create projects, set policies, assign users

2.

End User Portal: Deploy and monitor applications

Benefits:

• Frictionless service delivery
• Developers empowered to get AWS like

experience

Prism Pro – Advanced operations features

Anomaly detection and dynamic alerts VM Efficiency & Right- sizing Scheduled Reporting

Multi-Cloud Is The New Normal Today

Enterprise Applications Productivity Workloads AI, ML & Data Applications Burstable Workloads

Right Cloud for the right workload Laws of land, laws of physics and laws of locality De-risk reliance on single cloud for the business

Nutanix Confidential – DO NOT DISTRIBUTE

Complexity In Multi-Cloud Brings Many Challenges

Budgeting and Planning Finan ancia cial l Gover ernan nance ce For Teams ams Cost t Optimi imiza zatio tion n Acros

• ss

s Cloud uds Visib ibili ilit y Acros ross s Cloud uds Cloud ud Compl plia ianc nce Cloud ud Resou

• urce

rces s Spraw awl Consu sump mpti tion

• n

Cost t Contro trol Track ckin ing g Resou

• urce

rces Fine Grained Audit Controls Spira rali ling ng Cost sts

Beam and Calm: Cloud from Both Sides

Cost Optimization Security Compliance Provisioning and Lifecycle Operations Self-Service for End-Users

Multi-Cloud Governance

Top Down Business Challenges Bottom-up Operator Challenges

Multi-Cloud Application Management

Calm

Nutanix Era

NUTANIX ERA | CONFIDENTIAL

Multiple Engines GBs TBs Different Sizes

Monitori rin g Provision ionin in g LCM LCM

Complex Operations

Database Operations Complexities

Lots of Copies

Thousands

Traditional DB Provisioning Process

Specifies DB Config Gets DB Configure compute Allocate storage Creates a Cluster Setup Network Provisions DB Protect DB Requests DB

Time 0

NUTANIX ERA | CONFIDENTIAL

• Takes hours, days, or weeks
• Involves multiple teams
• Introduces friction & complex process
• Lowers business agility
• Decreases innovation rate
• Increases time to market

Nutanix Era – DBaaS for the Enterprise Cloud

NUTANIX ERA | CONFIDENTIAL

Not GA Roadmap

Volumes Files Buckets

Acropolis - HCI Platform

vSphere AHV

Infrastructure Services

Storage Services Virtualization

Recovery, Update - Patching/upgrade Time-machine Snap, Clone/Refresh, Backup Provisioning

Platform Services

Monitoring DR

Databases

API CLI GUI

Provisioning Snap, Clone Replication

Nutanix Karbon

Application Evolution

Traditional Applications

User Interface Business Logic Data Access Layer

Single Deployable Unit

• Complex code structure to maintain
• Update, modification, and upgrades are

complex operations

• Slow application code update
• Requires build & deployment of a new

version of applications for any code update

• Long application qualification time
• Decreases business Agility

Microservices

User Interface

Microservice Microservice

• Update, modification, and upgrades are simple operations
• Fast application code update
• Short application qualification time
• Increase business Agility

Microservice Microservice

T I T L E O F P R E S E N T A T I O N | C O N F I D E N T I A L

Containers are the Key Enabler

Heavyweight Each VM runs in its own OS Hardware-level virtualization Startup time in minutes Allocates required memory Fully isolated and hence more secure Less potable Lightweight All containers share the host OS OS virtualization Startup time in milliseconds Requires less memory space Process-level isolation, possibly less secure Extremely portable “By 2020, more than 50% of enterprises will run mission-critical, containerized cloud-native applications in production, up from less than 5% today.”

Nutanix Karbon

Managed Kubernetes offering for deploying and managing Kubernetes Clusters.

• Production ready multi-cluster Kubernetes management

platform.

• Prism user experience for Kubernetes to abstract

Kubernetes operational complexities

• Extends upstream Kubernetes capabilities, improving its

resilience and security.

• Integrated Storage, Networking, Logging & Monitoring.

One-Click Kubernetes in three basic steps

• Provision a Kubernetes cluster
• Setup the container network
• Setup the storage provider

Frame Desktop and Application Delivery NX Hardware Platform Flow Networking Acropolis Hyperconverged AHV ESXi Hyper-V Virtualization Platform and Infrastructure Services Era Buckets Files Karbon Volumes Prism Infrastructure Management

IoT Leap

Xi AWS Azure GCP Beam Calm Epoch App & Multi-Cloud Management

Xi Cloud Services

Enterprise Cloud with Freedom to Choose

T H E E N T E R P R I S E C L O U D | 59

Clouds

Demo

Modernize IT with HCI

Secure and automate applications, and consolidate storage

Build an Enterprise Cloud Multi-Cloud Services for Apps and Data

Nutanix Customer Journey

Unify operations across public and private clouds Deliver enterprise apps and VDI from any site

E N T E R P R I S E C L O U D | C O N F I D E N T I A L

Hypervisor Choice Platform Choice Cloud Choice

Freedom to Choose

Nutanix AHV More

| 62

Undisputed Leader

| 63 Challengers Leaders Niche Players Visionaries

Nutanix, 51 VMware, 41 Other, 8

Gartner Magic Quadrant for HCI, 2018 The Forrester Wave Hyper-Converged Infrastructure, 2018 IDC’s Competitive Landscape of HCI Software Only, 2018

Durham Amsterdam San Jose Sydney Bangalore Tokyo Beijing

Industry’s Best Support

| 64

NBD Depots Support Centers

Over 100 Countries

24x7x365

“Follow the Sun” Support Proactive Support with Pulse

<30 min Mission Critical

Support Response Time

7 WW Support Centers 97% Customer Satisfaction +90 Net Promoter Score

for > 4 Years

Thank You