1Z0-1072-25 Question 11
Single answerYou are designing a high-throughput data processing application that regularly writes and processes large volumes of short-lived (ephemeral) data. The team wants to optimize for low latency and high I/O performance, but they also need the flexibility to right-size the CPU core count and memory for changing workloads. Which compute shape in Oracle Cloud Infrastructure is the most appropriate choice?
- A
VM.DenseIO.E4.Flex
- B
HPC2.36
- C
VM.Standard.E4.Flex
- D
BM.GPU.A10
Show answer and explanation
Correct answer: A
Explanation
To handle large volumes of ephemeral data writes, DenseIO shapes in OCI come with local NVMe storage that offers very high I/O performance. Since the requirement also includes adjusting CPU and memory capacity with minimal overhead, the VM.DenseIO.E4.Flex shape is the best match. It combines local ephemeral storage for high-speed data reads/writes and flexible resource configurations. Refer to the official OCI documentation on Compute shapes and DenseIO configurations for detailed considerations and best practices.
- A. Correct.
Option 1 (VM.DenseIO.E4.Flex) is correct because DenseIO shapes offer local NVMe storage for high I/O performance on ephemeral data. The 'E4.Flex' aspect provides flexible configuration of OCPUs and memory, allowing you to optimize resource usage to match fluctuating workloads without committing to the cost of a full bare metal server.
- B. Incorrect.
Option 2 (HPC2.36) is designed for highly specialized high-performance computing (HPC) workloads, typically involving large-scale parallel computations. While it offers excellent compute performance for MPI workloads or large-scale simulations, it is not specifically targeted at frequent, short-lived data writes or the flexibility to right-size CPU/memory for general workloads.
- C. Incorrect.
Option 3 (VM.Standard.E4.Flex) gives flexibility for CPU and memory configuration but does not provide local NVMe storage for high I/O throughput. It primarily uses remote block storage, which is not ideal for a scenario requiring fast ephemeral data access.
- D. Incorrect.
Option 4 (BM.GPU.A10) is a bare metal shape with GPU capabilities. It is suited for GPU-accelerated workloads such as machine learning training or graphics rendering. It does not inherently address fast ephemeral storage needs nor does it offer the same level of flexible CPU/memory scaling for non-GPU workloads.