Soru

Zorluk: OrtaResilient and Highly Available Storage Infrastructure

An automotive engineering firm is designing a disaster recovery solution for its on-premises computer-aided design (CAD) file server, which holds 40 TB40\text{ TB} of data. The firm has defined a Recovery Point Objective (RPO) of 4 hours4\text{ hours} and a Recovery Time Objective (RTO) of 2 hours2\text{ hours}. During normal operations, on-premises users must access these files with sub-millisecond local latency. In the event of a local site failure, the files must be immediately accessible from a highly available and resilient environment in AWS. Which storage architecture meets these requirements most cost-effectively?

  1. Deploy AWS Storage Gateway in cached volume mode on-premises. Cache frequently accessed files locally, store the full dataset in Amazon S3, and schedule volume snapshots every 4 hours4\text{ hours}. In a disaster recovery event, restore the snapshots as Amazon Elastic Block Store (Amazon EBS) volumes and attach them to Amazon EC2 instances.Cevap
  2. B
    Deploy AWS Storage Gateway in stored volume mode on-premises. Replicate the local volumes to Amazon S3 Glacier Flexible Retrieval daily. In a disaster recovery event, use Standard retrieval to restore the archives to Amazon EBS volumes attached to Amazon EC2 instances.
  3. C
    Implement a Pilot Light disaster recovery strategy. Configure AWS Storage Gateway in file gateway mode on-premises. In the event of a disaster, spin up a scaled-down Amazon EC2 instance that mounts the S3 bucket, expecting it to serve the entire production workload immediately without any manual configuration or scaling steps.
  4. D
    Store the CAD files on an Amazon RDS database. Set up an RDS Read Replica in a different Availability Zone, and configure the application to automatically promote the read replica to the primary instance for immediate, automated write failover during an outage.

Cevap

Deploy AWS Storage Gateway in cached volume mode on-premises, storing the full dataset in Amazon S3, scheduling snapshots every 4 hours4\text{ hours}, and restoring these snapshots to Amazon EBS volumes attached to Amazon EC2 instances during a disaster.
The correct solution uses AWS Storage Gateway in cached volume mode to store the master dataset in Amazon S3 and cache active files locally for sub-millisecond access. Taking snapshots every 4 hours4\text{ hours} meets the 4-hour4\text{-hour} RPO, and restoring those snapshots to EBS volumes attached to EC2 instances in a disaster meets the 2-hour2\text{-hour} RTO.

Adım Adım Çözüm

1
Analyze the local latency and storage architecture requirements.
The requirement for sub-millisecond local latency on-premises indicates that frequently accessed data must be cached locally. AWS Storage Gateway in cached volume mode fulfills this by storing the primary dataset in Amazon S3 and maintaining a local cache of active data.
This minimizes local latency while ensuring durability by storing the master copy in the AWS cloud.
2
Evaluate the Recovery Point Objective (RPO) constraint.
An RPO of 4 hours4\text{ hours} means that backup or replication frequency must occur at least every 4 hours4\text{ hours} to prevent losing more than 4 hours4\text{ hours} of data.
Scheduling volume snapshots every 4 hours4\text{ hours} directly satisfies this requirement.
3
Evaluate the Recovery Time Objective (RTO) constraint.
An RTO of 2 hours2\text{ hours} means the system must be restored and running within 2 hours2\text{ hours}. Restoring EBS snapshots and attaching them to EC2 instances takes minutes, which is well within the 2-hour2\text{-hour} window.
Using Standard retrieval from Glacier Flexible Retrieval (which takes 33 to 5 hours5\text{ hours}) is ruled out because it cannot meet this RTO.

Anahtar Kavram

AWS Storage Gateway cached volumes provide low-latency local access to frequently used data while securing the full dataset in Amazon S3, allowing rapid disaster recovery via EBS snapshots.
Tahmini Süre:1m 30s
Bu soruyu puanla