MMDB cluster system is a memory optimized relation database that implements on cluster computing platform, provides applications with extremely fast response time and very high throughput as required by many applications in a wide range of industries. With data is memory resident and site's memory capacity is limited, proposed algorithm will help to find an effective data placement that is the key point of improving performance of MMDB clusters. Here, a new dynamic fragment allocation algorithm in Partially Replicated Allocation scenario is proposed. This algorithm reallocates data with respect to changing data access pattern for each fragment in which data is maintained in current site, migrated or created replicas on remote site depend on accessing frequency and average response time. Proposed algorithm is suitable for MMDB cluster because it provide a better response time and maximize the locality of processing so could develop parallel processing of MMDB in cluster environment.