Lately, memory bandwidth and latency have become the key reasons for slowing down the efficiency of mobile AI computing. To curb this issue, it is reported that Chinese manufacturers Xiaomi and Huawei are exploring a low-latency wide DRAM solution called LLW.
According to the AnTuTu report, LLW borrows from the high-integration design concept of HBM, but it has been redesigned for smartphones, avoiding the packaging and heat dissipation obstacles of HBM.
LLW is supposed to break through the bandwidth and latency limitations of traditional LPDDR through a tighter integration architecture. It is said that LLW can theoretically reduce power consumption by 50% and achieve a performance improvement of about 1.5 times, though the actual effect still needs to be verified.
LLW is expected to move data between the processor and memory much faster, reduce latency, and keep AI models fed with data instead of waiting on memory.
The mass adoption of LLW is still some time away, but Xiaomi and Huawei, with LLW-related products, are expected to see substantial progress in the second half of 2027. But the companies have not announced anything related to this till now.
Stay tuned with The Tech Outlook for the latest tech updates!
