<p>While SSD RAID configurations offer advantages in I/O parallelism and data redundancy, they can suffer from significant performance degradation under random write-intensive workloads, particularly those involving small I/O operations. More critically, the resulting uneven write distribution disrupts the synchronization of garbage collection (GC) processes across the SSDs, which leads to increased latency in subsequent I/O operations. To mitigate this issue, this article introduces a Small Writes Aggregate Cache (SAC) designed to coalesce random small writes into larger, more efficient units, thereby enhancing the overall parallel write efficiency of the SSD RAID. This scheme enables concurrent write operations across all SSDs and induces synchronized GC. To rapidly identify cacheable data suitable for forming full stripes, the system employs a stripe bitmap table to track the status of cached data. Moreover, we propose a stripe least recently used (LRU) algorithm to manage the cache space under constraints. When forming additional full stripes is infeasible, this algorithm uses the stripe bitmap table to efficiently evict partial stripe data across different devices. This approach ensures that partial stripe writes have an impact on GC operations comparable to that of full stripe writes. Through a series of simulation tests on several realistic disk workloads, we show that SAC effectively synchronizes GC operations across SSDs and reduces overall I/O response time compared to state-of-the-art PPC and Patch schemes.</p>