A retail company stores order invoices in an Amazon OpenSearch Service (Amazon Elasticsearch Service) cluster. Indices on the cluster are created monthly. Once a new month begins, no new writes are made to any of the indices from the previous months. The company has been expanding the storage on the Amazon OpenSearch Service (Amazon Elasticsearch Service) cluster to avoid running out of space, but the company wants to reduce costs. Most searches on the cluster are on the most recent 3 months of data, while the audit team requires infrequent access to older data to generate periodic reports. The most recent 3 months of data must be quickly available for queries, but the audit team can tolerate slower queries if the solution saves on cluster costs.
Which of the following is the MOST operationally efficient solution to meet these requirements?

A. Archive indices that are older than 3 months by using Index State Management (ISM) to create a policy to store the indices in Amazon S3 Glacier Instant Retrieval. When the audit team requires the archived data, restore the archived indices back to the Amazon OpenSearch Service (Amazon Elasticsearch Service) cluster.
B. Archive indices that are older than 3 months by taking manual snapshots and storing the snapshots in Amazon S3. When the audit team requires the archived data, restore the archived indices back to the Amazon OpenSearch Service (Amazon Elasticsearch Service) cluster.
C. Archive indices that are older than 3 months by using Index State Management (ISM) to create a policy to migrate the indices to Amazon OpenSearch Service (Amazon Elasticsearch Service) UltraWarm storage.
D. Archive indices that are older than 3 months by using Index State Management (ISM) to create a policy to migrate the indices to Amazon OpenSearch Service (Amazon Elasticsearch Service) UltraWarm storage. When the audit team requires the older data, migrate the indices in UltraWarm storage back to hot storage.