From: Loic Dachary <loic@dachary.org>
Date: Sun, 22 Sep 2013 08:58:23 +0000 (+0200)
Subject: ErasureCode: minimum_to_decode unit tests and optimization
X-Git-Tag: v0.71~98^2
X-Git-Url: http://git-server-git.apps.pok.os.sepia.ceph.com/?a=commitdiff_plain;h=9dfac9edfe4beedcd0c51e553c75411dd045685c;p=ceph.git

ErasureCode: minimum_to_decode unit tests and optimization

The minimum_to_decode function simply returns the available chunks if no
recovery is necessary.

Add unit tests covering all minimum_to_decode situations:

* trying to read nothing
* read a chunk if none are available
* reading a subset of the available chunks
* read a missing chunk if there is less than k chunks available.
* reading when a chunk must be recovered

Signed-off-by: Loic Dachary <loic@dachary.org>
---

diff --git a/src/osd/ErasureCodePluginJerasure/ErasureCodeJerasure.cc b/src/osd/ErasureCodePluginJerasure/ErasureCodeJerasure.cc
index 25821274081d3..aa22144bb6e56 100644
--- a/src/osd/ErasureCodePluginJerasure/ErasureCodeJerasure.cc
+++ b/src/osd/ErasureCodePluginJerasure/ErasureCodeJerasure.cc
@@ -15,6 +15,7 @@
  */
 
 #include <errno.h>
+#include <algorithm>
 #include "common/debug.h"
 #include "ErasureCodeJerasure.h"
 extern "C" {
@@ -43,12 +44,17 @@ void ErasureCodeJerasure::init(const map<std::string,std::string> &parameters) {
 int ErasureCodeJerasure::minimum_to_decode(const set<int> &want_to_read,
                                            const set<int> &available_chunks,
                                            set<int> *minimum) {
-  if (available_chunks.size() < (unsigned)k)
-    return -EIO;
-  set<int>::iterator i;
-  unsigned j;
-  for (i = available_chunks.begin(), j = 0; j < (unsigned)k; i++, j++)
-    minimum->insert(*i);
+  if (includes(available_chunks.begin(), available_chunks.end(),
+	       want_to_read.begin(), want_to_read.end())) {
+    *minimum = want_to_read;
+  } else {
+    if (available_chunks.size() < (unsigned)k)
+      return -EIO;
+    set<int>::iterator i;
+    unsigned j;
+    for (i = available_chunks.begin(), j = 0; j < (unsigned)k; i++, j++)
+      minimum->insert(*i);
+  }
   return 0;
 }
 
diff --git a/src/test/osd/TestErasureCodeJerasure.cc b/src/test/osd/TestErasureCodeJerasure.cc
index 266b1735659aa..22aaff7e5fac9 100644
--- a/src/test/osd/TestErasureCodeJerasure.cc
+++ b/src/test/osd/TestErasureCodeJerasure.cc
@@ -14,6 +14,7 @@
  * 
  */
 
+#include <errno.h>
 #include "global/global_init.h"
 #include "osd/ErasureCodePluginJerasure/ErasureCodeJerasure.h"
 #include "common/ceph_argparse.h"
@@ -36,7 +37,8 @@ typedef ::testing::Types<
 > JerasureTypes;
 TYPED_TEST_CASE(ErasureCodeTest, JerasureTypes);
 
-TYPED_TEST(ErasureCodeTest, encode_decode) {
+TYPED_TEST(ErasureCodeTest, encode_decode)
+{
   TypeParam jerasure;
   map<std::string,std::string> parameters;
   parameters["erasure-code-k"] = "2";
@@ -92,7 +94,106 @@ TYPED_TEST(ErasureCodeTest, encode_decode) {
   }
 }
 
-int main(int argc, char **argv) {
+TYPED_TEST(ErasureCodeTest, minimum_to_decode)
+{
+  TypeParam jerasure;
+  map<std::string,std::string> parameters;
+  parameters["erasure-code-k"] = "2";
+  parameters["erasure-code-m"] = "2";
+  parameters["erasure-code-w"] = "7";
+  parameters["erasure-code-packetsize"] = "8";
+  jerasure.init(parameters);
+
+  //
+  // If trying to read nothing, the minimum is empty.
+  //
+  {
+    set<int> want_to_read;
+    set<int> available_chunks;
+    set<int> minimum;
+
+    EXPECT_EQ(0, jerasure.minimum_to_decode(want_to_read,
+					    available_chunks,
+					    &minimum));
+    EXPECT_TRUE(minimum.empty());
+  }
+  //
+  // There is no way to read a chunk if none are available.
+  //
+  {
+    set<int> want_to_read;
+    set<int> available_chunks;
+    set<int> minimum;
+
+    want_to_read.insert(0);
+
+    EXPECT_EQ(-EIO, jerasure.minimum_to_decode(want_to_read,
+					       available_chunks,
+					       &minimum));
+  }
+  //
+  // Reading a subset of the available chunks is always possible.
+  //
+  {
+    set<int> want_to_read;
+    set<int> available_chunks;
+    set<int> minimum;
+
+    want_to_read.insert(0);
+    available_chunks.insert(0);
+
+    EXPECT_EQ(0, jerasure.minimum_to_decode(want_to_read,
+					    available_chunks,
+					    &minimum));
+    EXPECT_EQ(want_to_read, minimum);
+  }
+  //
+  // There is no way to read a missing chunk if there is less than k
+  // chunks available.
+  //
+  {
+    set<int> want_to_read;
+    set<int> available_chunks;
+    set<int> minimum;
+
+    want_to_read.insert(0);
+    want_to_read.insert(1);
+    available_chunks.insert(0);
+
+    EXPECT_EQ(-EIO, jerasure.minimum_to_decode(want_to_read,
+					       available_chunks,
+					       &minimum));
+  }
+  //
+  // When chunks are not available, the minimum can be made of any
+  // chunks. For instance, to read 1 and 3 below the minimum could be
+  // 2 and 3 which may seem better because it contains one of the
+  // chunks to be read. But it won't be more efficient than retrieving
+  // 0 and 2 instead because, in both cases, the decode function will
+  // need to run the same recovery operation and use the same amount
+  // of CPU and memory.
+  //
+  {
+    set<int> want_to_read;
+    set<int> available_chunks;
+    set<int> minimum;
+
+    want_to_read.insert(1);
+    want_to_read.insert(3);
+    available_chunks.insert(0);
+    available_chunks.insert(2);
+    available_chunks.insert(3);
+
+    EXPECT_EQ(0, jerasure.minimum_to_decode(want_to_read,
+					    available_chunks,
+					    &minimum));
+    EXPECT_EQ(2u, minimum.size());
+    EXPECT_EQ(0u, minimum.count(3));
+  }
+}
+
+int main(int argc, char **argv)
+{
   vector<const char*> args;
   argv_to_vec(argc, (const char **)argv, args);