A new semiconductor memory IP product provides an alternative to traditional multiport implementations, yielding power and area savings; reduce overall time-to-market and do not require silicon validation. The Renaissance 4X uses algorithms synthesized in hardware to increase the performance of existing embedded memory macros. The technology is implemented as RTL-level IP, and is process, node, and foundry independent. Memoir’s solutions appear as standard embedded memories, and can be readily integrated into existing standard SoC design flows.
Based upon the company’s Algorithmic Memory technology, Renaissance 4X offers six separate memory generators: a 3 port 2R1W, a 3 port 1R2W, a 4 port 3R1W, a 4 port 2R2W, and specialty memories for 3Ror1W and 4Ror1W. It is designed to increase the memory performance of existing embedded memory macros by delivering up to a 4X increase in memory operations per second (MOPS). In addition, it eliminates the need to build custom multiport memories and can reduce area and power requirements by up to 60% compared to conventional physical multiport implementations. By combining memory algorithms with silicon-validated single-port/two-port memories, Renaissance 4X can generate memories with any read/write combinations for up to four active ports, easily achieving 4000 MOPS in a 28-nm process.
Renaissance 4X alleviates the need for custom memory solutions and creates a versatile memory portfolio offering of six multiport memory generators that cater to different memory requirements and application needs. The product uses standard 6T BIST and DFT methodologies, contains RTL that has been exhaustively verified using formal methodologies, and does not require silicon validation. Through this combination of benefits, Renaissance 4X provides a significant time-to-market advantage, lowers product development cost, and reduces risk. With Renaissance 4X, customers have flexibility and options not available with custom memory solutions. For instance, memory replication is area inefficient, requires much higher power and only works to increase read performance. On the other hand, custom multiport solutions have higher costs, risks, and require a longer time to market. This means that they are not generally offered by third-party IP suppliers.
From an application standpoint, Renaissance 4X meets the data requirements of next-generation SoCs used in networking and communication sub-systems which have aggregated speeds above 400 Gb/s. This means that Renaissance 4X has extensive applicability for memories which require multiple memory accesses per cycle, such as packet buffers, counters, netflow, linked lists, schedulers, lookup tables, etc. In addition, Renaissance 4X provides multiport memories for shared L2 and L3 cache architectures, for high performance multicore SoCs.
Pricing and availability
Renaissance 4X Generators are available now and list pricing starts at $500,000 plus royalties.
Did you find this article of interest? Then visit the Memory Designline
where we update daily with design, technology, product, and news
articles tailored to fit your world. Too busy to go every day? Sign up
for our newsletter to get the week's best items delivered to your inbox.
Just click here
and choose the "Manage Newsletters" tab.