A financial services company is migrating a high-frequency trading simulation platform to Amazon EC2. The platform consists of a cluster of Linux EC2 instances that require concurrent, shared access to a dataset via the NFS protocol. The workload demands sub-millisecond latencies for random file operations, sustained throughput of during peak hours, and up to IOPS. The storage solution must also support instantaneous, space-efficient file system snapshots and clones for testing runs.
Which combination of storage configurations and client-side settings will meet these requirements? (Select TWO.)
- Provision an Amazon FSx for NetApp ONTAP file system with SSD storage, configuring the throughput capacity and provisioned IOPS to meet the workload demands.Cevap
- Mount the shared file system on the Linux EC2 instances using the NFS protocol with the 'nconnect' mount option enabled.Cevap
- CCreate an Amazon EFS file system configured in Max I/O performance mode and Provisioned throughput mode set to .
- DProvision an Amazon EBS gp3 volume and enable Multi-Attach to share the block storage across all EC2 instances in the cluster.
- EMount the file system using standard NFSv4 without the 'nconnect' mount option, relying on Amazon EC2 Enhanced Networking to automatically scale TCP connections.
Cevap
Provision an Amazon FSx for NetApp ONTAP file system with SSD storage and mount it using the NFS protocol with the 'nconnect' mount option enabled.
Provisioning an Amazon FSx for NetApp ONTAP file system is correct because it supports NFSv4, SSD storage with sub-millisecond latencies, and native features like snapshots and clones. Mounting the file system with the 'nconnect' option is correct because it enables the Linux NFS client to use multiple TCP connections, which is required to achieve the throughput of and IOPS that would otherwise be bottlenecked by a single TCP connection.
Adım Adım Çözüm
Anahtar Kavram
Amazon FSx for NetApp ONTAP offers sub-millisecond latencies, NFS compatibility, and advanced data management (snapshots/clones), which must be paired with client-side NFS tuning (such as the 'nconnect' mount option) to bypass single-TCP connection throughput limits.