Whitenoise is as fast in software as it is in hardware (microprocessors). Deployment through microprocessors will allow Whitenoise Technologies to be deployed in smart grids and critical infrastructures even though many of the components are cost sensitive, or are severely restricted. Manufacturers are already buying chip sets so simply swapping them out with Whitenoise enabled chip sets in the next product cycle represents zero cost for adoption - a smart move.
Integer factorization cryptography like RSA are difficult to implement and require a lot of overhead. And they are inefficient.
Dr. Rivest, one of the founders of RSA, announced an improvement on an algorithm at a talk at Charles Rive Cyber 2014:
"We estimate that Spritz can produce output with about 24 cycles/byte of computation. (sic 1 byte per 24 clock cycles). Furthermore, our statistical tests suggest that about $2^{81}$ bytes of output are needed before one can reasonably distinguish Spritz output from random output; this is a marked improvement over RC4."
Alternatively, Whitenoise can produce output with 2 bytes per clock cycle. And that is scalable in multi-threading/multi-core environments.
That is orders of magnitude faster than Spritz (minimum 48 times faster) with virtually no overhead or computational requirements. This illustrates exactly why RSA technologies and asymmetric processes are not a good choice for securing data centers and big data. This illustrates exactly why RSA cannot effectively deploy even 128 bit keys in the majority of devices that comprise the Internet of Things and the Cloud of Things.
Please download whitepaper about Whitenoise Technologies in microprocessors:
Whitenoise Deployed in Chips White Paper
Addendum
Characteristics of a Whitenoise key DATA SOURCE.
The data source [Subkey 1, 2 ...n] loops horizontally. It can only be used to the point where all the seams between all the subkeys lined up at which point it would lose a characteristic of a one-time-pad.
It does NOT operate horizontally which would make it a line feed shift register.
Rather each bit is X-Or'd with the corresponding bit of the next subkey vertically to create initial data source key stream. This is then pushed through an S-Box to delinearize it and we generate our key stream which can be used for any network security control. And, X-Or is the fastest function available on a computer so we are ALWAYS THE FASTEST FUNCTION AND ALWAYS FASTER THAN THE HACKERS.
The key stream is bit independent which means that there is no reframing with flipped bits. You can instantly jump anywhere into the key stream to encrypt or decrypt.
You can parse keys and keys streams to run channels in parallel to dramatically decrease transfer times and dramatically increase download speeds.
--
One Way Functions
The security of any crypto system is determined by the strength or security of its "one way functions." The security of Public Key Instrastructures is based on the belief that it is very easy to multiply primes together but infeasible to go the reverse order and find the factors. The mathematics entailed with this process dramatically hits overhead and computational resources and performance. It bloats data. RSA themselves discontinued factoring challenges years ago because computer speeds were making techniques like the sieve method of factorization simple. There are techniques to factor public keys instantly.
It is simple to understand the Whitenoise one-way functions that make it unbreakable:
- It is a one-time pad, provably unbreakable mathematically.
- The process is mechanical (so no overhead) and not mathematical. Two bytes are drawn from the initial key stream and appended together. This is pushed through the S-Box and only one byte emerges at Step 3. It is further X-Or with another by from the initiatl key stream and this becomes the first byte of the Delinearized Whitenoise Identity Management key. A hacker cannot go backward and guess two bytes of information from only one byte of captured data.
- The hacker has no way of knowing the number of subkeys that were used or their prime number length.
- The subkey lengths themselves are filled with random data (not 1,2,3,4... like a line feed shift register).
--
Historical problems of key distribution, key management and key distribution in large distributed key systems is solved.
There is a one to one relationship between the number of Whitenoise keys to manage and the number of endpoints. Historically you had to store the square i.e. a distributed key system with 10 endpoints required managing the square of that number or 100 keys. [Note: WN does however do 1:1, 1:Many and Many:1 communications.)
Big keys are better than small ones. In the above example, we only need to store or transmit the key structure (3,5,7...29) or Whitenoise key DNA to deterministically recreate a key stream over 100 billion bytes long. That is exponentialism.