Next Hop can help you with BCDR by:
- Organizing your data
- Deploying, maintaining and testing backup systems
- Simulating recovery procedures
“If it’s not in three places, it doesn’t exist.”
Backups
A backup is a pristine, complete copy of your data.
Aim to implement a 3-2-1 backup strategy:
- 3 . . . complete copies of the data
- 2 . . . different types of media
- 1 . . . copy kept off-site
All of the following concepts are part of a backup strategy, but are not backups on their own:
- Snapshots (VMFS, BTRFS, ZFS, Veeam, etc)
- On-site or cross-site replication (mirroring)
- High-availability (HA)
You’re a small business. Why do you need help with BCDR?
-
Your company is about to scale up and you need help with your IT and storage needs
-
You run a data-heavy business like video marketing, and all your data is stored on one NAS
-
You manage the estate and collected works of an artist and have long-term archival needs
Business continuity
Continuity of business operations are paramount. While your IT vendors and staff will have a hand in architecting and implementing your business continuity plans (BCPs), they are principally a business decision. When consulting with a business on continuity planning, we might ask things like:
- Does it make sense to store an extra, offline copy of this data on tape?
- Are your store managers trained on inventory management in the event of an I.T. outage?
- Have you planned to automate paper inventory ingest when the system is recovered?
- Have the disaster recovery plans been tested recently?
RTO and RPO
Recovery time objective (RTO) is your target time to recover from an unplanned outage or incident. RTO should be calculated based on many factors including staff availability, data volume, retrieval costs, infrastructure limitations, etc. Your continuity plans should account for a realistic RTO.
Recovery point objective (RPO) is the point in time to which data can be recovered. One hour ago? Three days ago? This objective will vary based on your industry and business volume. If your organization is large, you may need to establish different objectives for different datasets.
Disaster recovery (DR)
This is the step-by-step procedure necessary to recover to a known-good state.
This procedure must be tested and war-gamed.
Functioning BCDR
So with all that said, what does a functioning BCDR system look like?
Here’s an example of a storage system with a number of availability and integrity features:
- High availability
- Snapshots
- On-site or hot site replication
- Cloud-based recovery environment
- Cloud-based, multi-region backup
- Cold site for recovery
graph TB
subgraph Office [Local Office]
direction LR
dataPrimary1["Dataset (HA_1)"]
dataPrimary2["Dataset (HA_2)"]
dataPrimary1 <== Replication (weekly, daily, hourly, tuned per dataset) ==> dataPrimary2
end
subgraph Hotsite [Nearby Hot Site]
direction LR
nearbyBackup["Nearby backup server"]@{shape: } --> nearbyExtra["Snapshots, LTO tape"]
style nearbyExtra stroke-width:2px,stroke-dasharray: 5 5
style nearbyBackup stroke-width:2px,stroke:#f66
end
subgraph cold_Site [Cold Site]
coldSite["Cold spare server"]
end
subgraph cloud_Backup [Cloud Backup Service]
direction TB
cloudBackup1["Distant region"]@{shape: cloud} --> cloud1Snaps["Snapshots & retention policies"]
style cloud1Snaps stroke-width:2px,stroke-dasharray: 5 5
style cloudBackup1 stroke-width:2px,stroke:#f66
end
subgraph cloud_VM [Cloud VM]
direction TB
cloudVM["Replica AD DC"]@{shape: cloud} --> VMsnap["Whole-VM hourly snapshots"]
style VMsnap stroke-width:2px,stroke-dasharray: 5 5
style cloudVM stroke-width:2px,stroke:#f66
end
Office -- Workload-tuned pushes (one-way operation) --> Hotsite
Hotsite --> cloud_Backup --> cold_Site
Office ----> cloud_VM --> cold_Site
linkStyle 4,5,6,7,8 stroke:#f66,stroke-width:2px;
A system like this provides robust protection against ransomware, data corruption, natural disasters, theft, etc.
There are numerous, redundant routes to a recovery point.
