La Agencia de Proyectos de Investigaci\u00f3n Avanzada de Defensa de EE.UU. (DARPA) ha anunciado cuatro nuevos equipos de investigaci\u00f3n, incluido uno dirigido por Intel, que intentar\u00e1n hacer del cifrado FHE una realidad pr\u00e1ctica. FHE ha sido durante mucho tiempo una aspiraci\u00f3n para los expertos en cifrado: permitir el c\u00e1lculo, el an\u00e1lisis y otros usos de la informaci\u00f3n cifrada sin necesidad de descifrarla. Esto ayudar\u00eda a lograr un mejor equilibrio entre poder utilizar datos confidenciales en su m\u00e1xima extensi\u00f3n y minimizar el riesgo de exposici\u00f3n, dice DARPA.<\/p>\n
The US Defense Advanced Research Projects Agency (DARPA) has announced four new research teams including one led by Intel that will try to make Full Homomorphic Encryption (FHE) a practical reality.<\/p>\n
FHE has long been an aspiration for encryption\u00a0experts: enabling computation, analysis and other uses of encrypted information without actually needing to decrypt it. This would help to strike a better balance between being able to use sensitive data to its fullest extent and minimizing the risk of exposure, DARPA says.<\/p>\n
The problem up until now has been the computing power and time required to achieve this.<\/p>\n
\u201cA computation that would take a millisecond to complete on a standard laptop would take\u00a0weeks to compute on a conventional server running FHE today,\u201d argued DARPA program manager, Tom Rondeau.<\/p>\n
To speed-up this processing time from weeks to seconds or milliseconds, DARPA is hoping to build a hardware accelerator as part of its Data Protection in Virtual Environments (DPRIVE) program, which would in theory offer major advances over software-based approaches.<\/p>\n
The research teams announced by DARPA yesterday are Intel\u2019s government-focused subsidiary Intel Federal, Duality Technologies, Galois and non-profit SRI International.<\/p>\n
The job of each is to create an FHE accelerator hardware and software stack designed to process FHE calculations at a similar speed to unencrypted data operations.<\/p>\n
In so doing, they will explore the use of CPUs with different sizes of \u201cwords\u201d \u2013\u00a0\u00a0the units of data that determine a processor\u2019s design. They\u2019ll try everything from the 64-bit words used in modern processor designs to 1000 bits.<\/p>\n
They\u2019ll also be looking into \u201cnovel approaches to memory management, flexible data structures and programming models, and formal verification methods,\u201d according to DARPA.<\/p>\n
If they\u2019re successful, it could have significant military and commercial applications.<\/p>\n
\u201cWe currently estimate we are about a million-times slower to compute in the FHE world then we are in the plaintext world,\u201d concluded Rondeau.<\/a><\/p>\n \u201cThe goal of DPRIVE is to bring FHE down to the computational speeds we see in plaintext. If we are able to achieve this goal while positioning the technology to scale, DPRIVE will have a significant impact on our ability to protect and preserve data and user privacy.\u201d<\/p>\n<\/div>\n<\/div>\n