Hybrid Cloud and SDDC Conceptual Design

Hybrid Cloud  and SDDC Conceptual DesignWorkload Profiles

  • VDI
  • Database Applications
  • Responsive  Apps
  • General Purpose

Performance Matrix

(Identify where you workload range and design solutions)

  • Compute Low, Medium, High
  • Memory
  • IOPS
  • Storage
  • Networking

Workload Categories

(Networking should be 10 GB or 40 GB infiBand)

  • General Purpose
    • SSD Cache
    • SATA
  • Compute Optimised
    • Latest Intel
    • SSD
    • SAS
  • Memory Optimised – For large databases, SharePoint server farms, and high-throughput applications
    • MAX Memory
    • SSD
    • SAS
  • Storage Optimised
    • FlashIO
    • SSD
    • Large SAS
  • Dataware Warehouse / File Servers
    • Large SAS
    • SMB3/CIFS/NFS

Instances Types (Match Azure)

Networks

  • DMZ
  • PCI
  • WIBs
  • Server
  • Desktop
  • Management

Design Decision

  • 100% Microsoft Cloud, StoreSimple and SuperMicro Server Solutions
  • Hybrid Backup and DRaaS Veeam No SAN Solution
  • VMware vSphere 6
  • FileShares on Azure Steelhead
  • Cloud IaaS (Azure, AWS) are not mature enough for complex Enterprise Workloads and Networking
  • Build to move all workloads to cloud in 3-5 years (Azure or VMware Air)
  • Veeam DR Partner
  • Virtualise Firewall
    • Fortinet or Checkpoint ASA
  • BCP Site
  • Daily Backups
    • Veeam Backup Partner
    • Azure Stheelhead
  • Long Term 7 Year Retention
    • AWS Storage Gateway Announces Gateway-Virtual Tape Library (Gateway-VTL)
    • Microsoft Data Protection to Azure Backup
    • Steelhead Azure WAN Accelerators
  • Co-lo (Managed Firewall and Switch)
  • Use Megaport NaaS – http://www.megaport.com/ecosystem/connected-data-centres.html

Capacity uplifts Standard

What is required to maintain and monitor capacity and availability?

Requirements

  • Windows VSS Enablement

DC

  • NextDC / CloudPlus and Azure ExpressRoute for Veeam Backup and / DR
  • Equinix / Nexon and Azure ExpressRoute

BCP & DR

  • Cloud repositories – use the new Veeam Cloud Connect functionality in v8 to get backups offsite easily and efficiently
  • Offsite replica VMs – maintain VM replicas in the cloud for quick recovery in the event of a disaste
  • Backup-as-a-Service (BaaS) powered by Veeam
    Offload backup to a Veeam Cloud Provider, and free up IT resources for more strategic business-building activities. You continue to run your VMs onsite, while your Veeam Cloud Provider provides backup to meet your RTOs and RPOs using the #1 VM Backup, Veeam Backup & Replication.
  • Disaster Recovery-as-a-Service (DRaaS) powered by Veeam
    Use a Veeam Cloud Provider to replicate your VMs to the cloud for fast recovery in the event of a disaster. You continue to backup VMs as you do today, while your Veeam Cloud Provider provides an added measure of protection by replicating VMs offsite.
  • Note: With BaaS and DRaaS, backup and replication are managed by the service provider (it’s not Veeam Backup & Replication delivered in a Software-as-a-Service, or SaaS model.)
  • http://www.veeam.com/find-a-veeam-cloud-provider.html

Right Sizing Information Gathering

  • VMware Capacity Planner or Microsoft Assessment and Planning Tool
  • NetAp nSanity AutoSupport Reports
  • Veeam One (VMware change)
  • Core Switch Performance Obsivrium
  • MAX/Averages (CPU, Networking, IOPS, Memory)
  • Tape Size
  • Backup Frequency and Sizes
  • RVTools

Compute, Networking and Memory Requirements

  • Storage Total Size 20 TB
  • CPU Mark Total
  • Memory Total
  • Backup Delta / Hot and Cold
  • Growth
  • CIFS Storage
  • Mail
  • SQL DBs
  • Core Switch Bandwidth (Business hours)
    • Average
    • Min/Max

Backup Retention Requirements

No-SAN Issues

  • Monolith VMs (Vertical Scale)
    • Large Memory Requirements
    • Large Storage Requirements
  • Clustering (All Ways On)
  • How to maintain capacity growth
  • Native CIFS/NFS
  • Separate DMZ ( need to use virtual networking)
    • http://www.cisco.com/c/en/us/products/security/asa-1000v-cloud-firewall/index.html
    • NSX

Components

  • SeaMicro
  • VMWare RAIL – http://www.vmware.com/au/products/evorail
  • Nutanix
  • VMWare AIR
  • Veeam
  • HP Converged
  • HP VSA
  • AppVolumes
  • VMware vSAN
  • Azure Inmage

Certified Hyper-V Compute and 10 GB Converged Networking

VM Templates

  • Small 1 vCPU, 2 GB memory, 50-GB disk
  • Med 2 vCPU, 4 GB memory,100-GB disk
  • Large 4 vCPU, 8 GB memory, 200-GB disk

Sizing Tools

  • http://h71019.www7.hp.com/ActiveAnswers/us/en/sizers/hp-sizer-server-virtualization.html
  • http://www.cisco.com/c/en/us/solutions/data-center-virtualization/data_center_value_zone.html
  • http://virtualsansizing.vmware.com/
  • Data Center Cost
  • http://www.cisco.com/c/en/us/support/web/tools-catalog.html
  • http://www.joshodgers.com/vsphere-cluster-sizing-calculator/

NetApp Performance

  • COLLECT AUTOSUPPORT FILES

1. Browse to the file share \\FILER_NAME\ETC$\log\autosupport
2. Locate the most recent folder with name YYMMDDHHMM.0 or .1
3. Ensure you have selected the most recent folder with a *.0 or .1 name.
4. Folder should have at least 150 files in it.
5. Package/zip the entire contents of this folder to provide to TD

  • COLLECT PERFORMANCE STATS FILES

1. Browse to the file share \\FILER_NAME\ETC$\log\stats\archive
2. Each *.gz file represents 1 hour of data
3. By default, collect 50 *.gz files (approximately 2 days) of recent activity that represents the typical workload on your filer. You can extend this to 100 files to assess a longer time period if desired.
4. Note: for ONTAP release 7.x browse to \\FILER_NAME\ETC$\stats\archive
5. Zip the Autosupport package with the performance stats package into a single compressed file and send to TD
6. Alternatively, you can forward an autosupport email

Cost Items

  • Azure Costs
    • VPN and Express Connect
    • Backup VM Storage
  • Co-Location
    • Switch
    • FC Switch
    • Internet Connections
  • Design Items
    • VMware
      • vSAN
      • DMZ (NSX/Checkpoint)

Critical Path

  • Costs Solution
  • Engage Microsoft Consulting Services to Validate Design
  • Acquire Rack Spaces and Servers/Equipment
  • Design IP and Networking Layer
  • Configure Routers and Layer 2 Bridge and ExpressRoute and MegaPort
  • Design VM, Server and Hybrid Solution
  • Test Components
  • Migrate Workloads to Equinix (Nexon) Production
  • Migrate Workloads to NextDC DC and setup UAT/DEV
  • Consolidate Workloads to Hyper-V
  • Integrate BackupExcu Tape Recovery

Reference

Migrate to Cloud

  • Cloud Security Policy (SANs Controls + SLA + Monitoring + Encryption + Access + DR + Restore)
  • Azure Backup
  • Extend Network to Azure Virtual Network
  • Azure VMs
  • Azure Site Recovery
  • Microsoft Migration Accelerator
  • Azure Network Extension
  • Application Dependency Mapping
  • DEV/UAT
  • SQL DB
  • File Server
  • Exchange DB
  • Sharepoint
  • Azure Web Sites
  • Responsive Web Servers (Full Solution)
  • Active Directory/2-Factor
  • Managed Firewall
  • Managed Switch
  • Visual Studio Online
  • Sharepoint
  • Load Balancer
  • Docker Apps Apps
  • Develop Cloud Security Policy

Leave a comment