VeraCrypt
aboutsummaryrefslogtreecommitdiff
path: root/doc/html/Parallelization.html
blob: 9da5d27b460dd0748fe9a3b0b563a8e1f736966e (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8" />
<title>VeraCrypt - Free Open source disk encryption with strong security for the Paranoid</title>
<meta name="description" content="VeraCrypt is free open-source disk encryption software for Windows, Mac OS X and Linux. In case an attacker forces you to reveal the password, VeraCrypt provides plausible deniability. In contrast to file encryption, data encryption performed by VeraCrypt is real-time (on-the-fly), automatic, transparent, needs very little memory, and does not involve temporary unencrypted files."/>
<meta name="keywords" content="encryption, security"/>
<link href="styles.css" rel="stylesheet" type="text/css" />
</head>
<body>

<div>                      
<a href="https://www.veracrypt.fr/en/Home.html"><img src="VeraCrypt128x128.png" alt="VeraCrypt"/></a>
</div>

<div id="menu">
	<ul>
	  <li><a href="Home.html">Home</a></li>
	  <li><a href="/code/">Source Code</a></li>
	  <li><a href="Downloads.html">Downloads</a></li>
	  <li><a class="active" href="Documentation.html">Documentation</a></li>
	  <li><a href="Donation.html">Donate</a></li>
	  <li><a href="https://sourceforge.net/p/veracrypt/discussion/" target="_blank">Forums</a></li>
	</ul>
</div>

<div>
<p>
<a href="Documentation.html">Documentation</a>           
<img src="arrow_right.gif" alt=">>" style="margin-top: 5px">
<a href="Parallelization.html">Parallelization</a>
</p></div>

<div class="wikidoc">
<h1>Parallelization</h1>
<div style="text-align:left; margin-top:19px; margin-bottom:19px; padding-top:0px; padding-bottom:0px">
When your computer has a multi-core processor (or multiple processors), VeraCrypt uses all of the cores (or processors) in parallel for encryption and decryption. For example, when VeraCrypt is to decrypt a chunk of data, it first splits the chunk into several
 smaller pieces. The number of the pieces is equal to the number of the cores (or processors). Then, all of the pieces are decrypted in parallel (piece 1 is decrypted by thread 1, piece 2 is decrypted by thread 2, etc). The same method is used for encryption.</div>
<div style="text-align:left; margin-top:19px; margin-bottom:19px; padding-top:0px; padding-bottom:0px">
So if your computer has, for example, a quad-core processor, then encryption and decryption are four times faster than on a single-core processor with equivalent specifications (likewise, they are twice faster on dual-core processors, etc).</div>
<div style="text-align:left; margin-top:19px; margin-bottom:19px; padding-top:0px; padding-bottom:0px">
Increase in encryption/decryption speed is directly proportional to the number of cores and/or processors.</div>
<div style="text-align:left; margin-top:19px; margin-bottom:19px; padding-top:0px; padding-bottom:0px">
Note: Processors with the Hyper-Threading technology provide multiple logical cores per one physical core (or multiple logical processors per one physical processor). When Hyper Threading is enabled in the computer firmware (e.g. BIOS) settings, VeraCrypt creates
 one thread for each logical core/processor. For example, on a 6-core processor that provides two logical cores per one physical core, VeraCrypt uses 12 threads.</div>
<p><br style="text-align:left">
When your computer has a multi-core processor/CPU (or multiple processors/CPUs), <a href="Header%20Key%20Derivation.html" style="text-align:left; color:#0080c0; text-decoration:none.html">
header key derivation</a> is parallelized too. As a result, mounting of a volume is several times faster on a multi-core processor (or multi-processor computer) than on a single-core processor (or a single-processor computer) with equivalent specifications.</p>
</div><div class="ClearBoth"></div></body></html>
} /* do length NMAX blocks -- requires just one modulo operation */ while (len >= NMAX) { len -= NMAX; n = NMAX / 16; /* NMAX is divisible by 16 */ do { DO16(buf); /* 16 sums unrolled */ buf += 16; } while (--n); MOD(adler); MOD(sum2); } /* do remaining bytes (less than NMAX, still just one modulo) */ if (len) { /* avoid modulos if none remaining */ while (len >= 16) { len -= 16; DO16(buf); buf += 16; } while (len--) { adler += *buf++; sum2 += adler; } MOD(adler); MOD(sum2); } /* return recombined sums */ return adler | (sum2 << 16); } /* ========================================================================= */ uLong ZEXPORT adler32(adler, buf, len) uLong adler; const Bytef *buf; uInt len; { return adler32_z(adler, buf, len); } /* ========================================================================= */ local uLong adler32_combine_(adler1, adler2, len2) uLong adler1; uLong adler2; z_off64_t len2; { unsigned long sum1; unsigned long sum2; unsigned rem; /* for negative len, return invalid adler32 as a clue for debugging */ if (len2 < 0) return 0xffffffffUL; /* the derivation of this formula is left as an exercise for the reader */ MOD63(len2); /* assumes len2 >= 0 */ rem = (unsigned)len2; sum1 = adler1 & 0xffff; sum2 = rem * sum1; MOD(sum2); sum1 += (adler2 & 0xffff) + BASE - 1; sum2 += ((adler1 >> 16) & 0xffff) + ((adler2 >> 16) & 0xffff) + BASE - rem; if (sum1 >= BASE) sum1 -= BASE; if (sum1 >= BASE) sum1 -= BASE; if (sum2 >= ((unsigned long)BASE << 1)) sum2 -= ((unsigned long)BASE << 1); if (sum2 >= BASE) sum2 -= BASE; return sum1 | (sum2 << 16); } /* ========================================================================= */ uLong ZEXPORT adler32_combine(adler1, adler2, len2) uLong adler1; uLong adler2; z_off_t len2; { return adler32_combine_(adler1, adler2, len2); } uLong ZEXPORT adler32_combine64(adler1, adler2, len2) uLong adler1; uLong adler2; z_off64_t len2; { return adler32_combine_(adler1, adler2, len2); }