M.2 storage drive format has two problems. The first refers to the thermal problems that these storage units present. In addition, we found a stagnation of capacity due to the space limitations of this very compact format.
Solution to capacity problem?
We must keep in mind that M.2 SSDs are a set of aspects that must be balanced. Storage capacity, operating speed, life, and cost must all go hand in hand. Achieving a balance between these aspects is not easy.
For these storage solutions are often used NAND Flash memory in quality SLC, MLC and TLC. But there QLC memory They have not yet taken off in this market, although they offer greater capacity. SLC memory stores 1 bit per cell, MLC stores 2 bits per cell, TLC 3 bits per cell, and QLC 4 bits per cell.
Kioxia, former Toshiba, would develop some new memory called HLC. These memories are characterized by storage 7 bits per cell, which is more than double the TLC memory. They would be the solution to significantly increase the capacity of SSD in the M.2 format.
The big particular memory of the HLC Kioxia is in the manufacturing. Apparently, it makes use of monocrystalline silicon, instead of the polycrystalline silicon that is often used in industry. This makes it possible to significantly increase the capacity.
Increase the amount of bits stored for each assumed cell reduce the price in the final product. This is because we would see the amount of memory on the PCB reduced, increasing the capacity at the same time. This means a reduction in manufacturing costs.
We could see a big increase in the capacity of M.2 SSDs in the future. It would dramatically increase the number of 8TB drives and we should see 16TB drives, no problem.
Predictably, we would see SSDs up to 16 TB
Thanks to this increase in bits per cell, a large increase in capacity should be achieved. An estimate, of M.2 format drives, could be between 16TB and 24TB of capacity. This would break the current ceiling of 4 TB.
At the same time that the brute capacity increases, the prices will be reduced. We could see 4TB drives going for around $100 for sure. In addition, the minimum capacity of an M.2 SSD would go from the current 256 GB to 1 TB.
We must say that these are estimates, since Kioxia has provided very little data. We do not know under what process these memories are manufactured, when they will be available or their benefits.
High cost to increase capacity
We all want more capacity, but this has high costs. The two main ones are: reduction in read/write speed and life.
The more the capacity increases, the more these two parameters are reduced. 1-bit per cell SLC memories are the fastest and have the longest life, but have been discontinued. The same thing happened with 2 bits per cell MLC, which despite being worse than the SLC, were great.
TLC (3 Bits) memory is now used for higher quality products. QLC (4 bits) has been relegated to low cost and “low” speed products. The same thing happens with the useful life of the units. The more the capacity grows, the lower the durability of the units.
HLC is still very green and it will take a long time to see them on the market. Assume new algorithms will be developed to prevent performance loss and advanced bug fixes. We should also see an increase in cache capacity. The loss of speed, in the end, is relative if we compare it with HDDs.