From eaff913d28ab736d517fb4df2488ecb563fd2fc2 Mon Sep 17 00:00:00 2001
From: Solomon Duskis <sduskis@google.com>
Date: Tue, 6 Jun 2017 09:47:40 -0400
Subject: [PATCH] Upgrading googlebigtable to the latest version. The API used
by googlebigtable has had quite a bit of churn. This is the minimal set of
changes required for the upgrade.
Signed-off-by: Chris Larsen <clarsen@yahoo-inc.com>
---
googlebigtable/README.md | 23 ++--
googlebigtable/pom.xml | 8 +-
.../yahoo/ycsb/db/GoogleBigtableClient.java | 100 ++++++++++--------
pom.xml | 2 +-
4 files changed, 72 insertions(+), 61 deletions(-)
diff --git a/googlebigtable/README.md b/googlebigtable/README.md
index 3938b525..81b6cf48 100644
--- a/googlebigtable/README.md
+++ b/googlebigtable/README.md
@@ -21,9 +21,9 @@ This driver provides a YCSB workload binding for Google's hosted Bigtable, the i
## Quickstart
-### 1. Setup a Bigtable Cluster
+### 1. Setup a Bigtable Instance
-Login to the Google Cloud Console and follow the [Creating Cluster](https://cloud.google.com/bigtable/docs/creating-cluster) steps. Make a note of your cluster name, zone and project ID.
+Login to the Google Cloud Console and follow the [Creating Instance](https://cloud.google.com/bigtable/docs/creating-instance) steps. Make a note of your instance ID and project ID.
### 2. Launch the Bigtable Shell
@@ -40,29 +40,25 @@ hbase(main):002:0> create 'usertable', 'cf', {SPLITS => (1..n_splits).map {|i| "
Make a note of the column family, in this example it's `cf``.
-### 4. Fetch the Proper ALPN Boot Jar
-
-The Bigtable protocol uses HTTP/2 which requires an ALPN protocol negotiation implementation. On JVM instantiation the implementation must be loaded before attempting to connect to the cluster. If you're using Java 7 or 8, use this [Jetty Version Table](http://www.eclipse.org/jetty/documentation/current/alpn-chapter.html#alpn-versions) to determine the version appropriate for your JVM. (ALPN is included in JDK 9+). Download the proper jar from [Maven](http://search.maven.org/#search%7Cgav%7C1%7Cg%3A%22org.mortbay.jetty.alpn%22%20AND%20a%3A%22alpn-boot%22) somewhere on your system.
-
-### 5. Download JSON Credentials
+### 4. Download JSON Credentials
Follow these instructions for [Generating a JSON key](https://cloud.google.com/bigtable/docs/installing-hbase-shell#service-account) and save it to your host.
-### 6. Load a Workload
+### 5. Load a Workload
-Switch to the root of the YCSB repo and choose the workload you want to run and `load` it first. With the CLI you must provide the column family, cluster properties and the ALPN jar to load.
+Switch to the root of the YCSB repo and choose the workload you want to run and `load` it first. With the CLI you must provide the column family and instance properties to load.
```
-bin/ycsb load googlebigtable -p columnfamily=cf -p google.bigtable.project.id=<PROJECT_ID> -p google.bigtable.cluster.name=<CLUSTER> -p google.bigtable.zone.name=<ZONE> -p google.bigtable.auth.service.account.enable=true -p google.bigtable.auth.json.keyfile=<PATH_TO_JSON_KEY> -jvm-args='-Xbootclasspath/p:<PATH_TO_ALPN_JAR>' -P workloads/workloada
+bin/ycsb load googlebigtable -p columnfamily=cf -p google.bigtable.project.id=<PROJECT_ID> -p google.bigtable.instance.id=<INSTANCE> -p google.bigtable.auth.json.keyfile=<PATH_TO_JSON_KEY> -P workloads/workloada
```
-Make sure to replace the variables in the angle brackets above with the proper value from your cluster. Additional configuration parameters are available below.
+Make sure to replace the variables in the angle brackets above with the proper value from your instance. Additional configuration parameters are available below.
The `load` step only executes inserts into the datastore. After loading data, run the same workload to mix reads with writes.
```
-bin/ycsb run googlebigtable -p columnfamily=cf -p google.bigtable.project.id=<PROJECT_ID> -p google.bigtable.cluster.name=<CLUSTER> -p google.bigtable.zone.name=<ZONE> -p google.bigtable.auth.service.account.enable=true -p google.bigtable.auth.json.keyfile=<PATH_TO_JSON_KEY> -jvm-args='-Xbootclasspath/p:<PATH_TO_ALPN_JAR>' -P workloads/workloada
+bin/ycsb run googlebigtable -p columnfamily=cf -p google.bigtable.project.id=<PROJECT_ID> -p google.bigtable.instance.id=<INSTANCE> -p google.bigtable.auth.json.keyfile=<PATH_TO_JSON_KEY> -P workloads/workloada
```
@@ -72,8 +68,7 @@ The following options can be configured using CLI (using the `-p` parameter) or
* `columnfamily`: (Required) The Bigtable column family to target.
* `google.bigtable.project.id`: (Required) The ID of a Bigtable project.
-* `google.bigtable.cluster.name`: (Required) The name of a Bigtable cluster.
-* `google.bigtable.zone.name`: (Required) Zone where the Bigtable cluster is running.
+* `google.bigtable.instance.id`: (Required) The name of a Bigtable instance.
* `google.bigtable.auth.service.account.enable`: Whether or not to authenticate with a service account. The default is true.
* `google.bigtable.auth.json.keyfile`: (Required) A service account key for authentication.
* `debug`: If true, prints debug information to standard out. The default is false.
diff --git a/googlebigtable/pom.xml b/googlebigtable/pom.xml
index 61eb48c4..7353e53a 100644
--- a/googlebigtable/pom.xml
+++ b/googlebigtable/pom.xml
@@ -36,6 +36,12 @@ LICENSE file.
<version>${googlebigtable.version}</version>
</dependency>
+ <dependency>
+ <groupId>io.netty</groupId>
+ <artifactId>netty-tcnative-boringssl-static</artifactId>
+ <version>1.1.33.Fork26</version>
+ </dependency>
+
<dependency>
<groupId>com.yahoo.ycsb</groupId>
<artifactId>core</artifactId>
@@ -44,4 +50,4 @@ LICENSE file.
</dependency>
</dependencies>
-</project>
\ No newline at end of file
+</project>
diff --git a/googlebigtable/src/main/java/com/yahoo/ycsb/db/GoogleBigtableClient.java b/googlebigtable/src/main/java/com/yahoo/ycsb/db/GoogleBigtableClient.java
index d0d21dda..c035032c 100644
--- a/googlebigtable/src/main/java/com/yahoo/ycsb/db/GoogleBigtableClient.java
+++ b/googlebigtable/src/main/java/com/yahoo/ycsb/db/GoogleBigtableClient.java
@@ -34,23 +34,24 @@ import java.util.Vector;
import java.util.concurrent.ExecutionException;
import com.google.bigtable.repackaged.com.google.protobuf.ByteString;
-import com.google.bigtable.repackaged.com.google.protobuf.ServiceException;
-import com.google.bigtable.v1.Column;
-import com.google.bigtable.v1.Family;
-import com.google.bigtable.v1.MutateRowRequest;
-import com.google.bigtable.v1.Mutation;
-import com.google.bigtable.v1.ReadRowsRequest;
-import com.google.bigtable.v1.Row;
-import com.google.bigtable.v1.RowFilter;
-import com.google.bigtable.v1.RowRange;
-import com.google.bigtable.v1.Mutation.DeleteFromRow;
-import com.google.bigtable.v1.Mutation.SetCell;
-import com.google.bigtable.v1.RowFilter.Chain.Builder;
+import com.google.bigtable.v2.Column;
+import com.google.bigtable.v2.Family;
+import com.google.bigtable.v2.MutateRowRequest;
+import com.google.bigtable.v2.Mutation;
+import com.google.bigtable.v2.ReadRowsRequest;
+import com.google.bigtable.v2.Row;
+import com.google.bigtable.v2.RowFilter;
+import com.google.bigtable.v2.RowRange;
+import com.google.bigtable.v2.RowSet;
+import com.google.bigtable.v2.Mutation.DeleteFromRow;
+import com.google.bigtable.v2.Mutation.SetCell;
+import com.google.bigtable.v2.RowFilter.Chain.Builder;
import com.google.cloud.bigtable.config.BigtableOptions;
import com.google.cloud.bigtable.grpc.BigtableDataClient;
import com.google.cloud.bigtable.grpc.BigtableSession;
+import com.google.cloud.bigtable.grpc.BigtableTableName;
import com.google.cloud.bigtable.grpc.async.AsyncExecutor;
-import com.google.cloud.bigtable.grpc.async.HeapSizeManager;
+import com.google.cloud.bigtable.grpc.async.BulkMutation;
import com.google.cloud.bigtable.hbase.BigtableOptionsFactory;
import com.google.cloud.bigtable.util.ByteStringer;
import com.yahoo.ycsb.ByteArrayByteIterator;
@@ -89,7 +90,6 @@ public class GoogleBigtableClient extends com.yahoo.ycsb.DB {
/** Thread loacal Bigtable native API objects. */
private BigtableDataClient client;
- private HeapSizeManager heapSizeManager;
private AsyncExecutor asyncExecutor;
/** The column family use for the workload. */
@@ -105,13 +105,21 @@ public class GoogleBigtableClient extends com.yahoo.ycsb.DB {
*/
private boolean clientSideBuffering = false;
+ private BulkMutation bulkMutation;
+
@Override
public void init() throws DBException {
Properties props = getProperties();
// Defaults the user can override if needed
- CONFIG.set("google.bigtable.auth.service.account.enable", "true");
-
+ if (getProperties().containsKey(ASYNC_MUTATOR_MAX_MEMORY)) {
+ CONFIG.set(BigtableOptionsFactory.BIGTABLE_BUFFERED_MUTATOR_MAX_MEMORY_KEY,
+ getProperties().getProperty(ASYNC_MUTATOR_MAX_MEMORY));
+ }
+ if (getProperties().containsKey(ASYNC_MAX_INFLIGHT_RPCS)) {
+ CONFIG.set(BigtableOptionsFactory.BIGTABLE_BULK_MAX_ROW_KEY_COUNT,
+ getProperties().getProperty(ASYNC_MAX_INFLIGHT_RPCS));
+ }
// make it easy on ourselves by copying all CLI properties into the config object.
final Iterator<Entry<Object, Object>> it = props.entrySet().iterator();
while (it.hasNext()) {
@@ -143,14 +151,7 @@ public class GoogleBigtableClient extends com.yahoo.ycsb.DB {
}
if (clientSideBuffering) {
- heapSizeManager = new HeapSizeManager(
- Long.parseLong(
- getProperties().getProperty(ASYNC_MUTATOR_MAX_MEMORY,
- Long.toString(AsyncExecutor.ASYNC_MUTATOR_MAX_MEMORY_DEFAULT))),
- Integer.parseInt(
- getProperties().getProperty(ASYNC_MAX_INFLIGHT_RPCS,
- Integer.toString(AsyncExecutor.MAX_INFLIGHT_RPCS_DEFAULT))));
- asyncExecutor = new AsyncExecutor(client, heapSizeManager);
+ asyncExecutor = session.createAsyncExecutor();
}
}
@@ -169,6 +170,13 @@ public class GoogleBigtableClient extends com.yahoo.ycsb.DB {
@Override
public void cleanup() throws DBException {
+ if (bulkMutation != null) {
+ try {
+ bulkMutation.flush();
+ } catch(RuntimeException e){
+ throw new DBException(e);
+ }
+ }
if (asyncExecutor != null) {
try {
asyncExecutor.flush();
@@ -226,7 +234,8 @@ public class GoogleBigtableClient extends com.yahoo.ycsb.DB {
final ReadRowsRequest.Builder rrr = ReadRowsRequest.newBuilder()
.setTableNameBytes(ByteStringer.wrap(lastTableBytes))
.setFilter(filter)
- .setRowKey(ByteStringer.wrap(key.getBytes()));
+ .setRows(RowSet.newBuilder()
+ .addRowKeys(ByteStringer.wrap(key.getBytes())));
List<Row> rows;
try {
@@ -292,13 +301,17 @@ public class GoogleBigtableClient extends com.yahoo.ycsb.DB {
}
final RowRange range = RowRange.newBuilder()
- .setStartKey(ByteStringer.wrap(startkey.getBytes()))
+ .setStartKeyClosed(ByteStringer.wrap(startkey.getBytes()))
.build();
-
+
+ final RowSet rowSet = RowSet.newBuilder()
+ .addRowRanges(range)
+ .build();
+
final ReadRowsRequest.Builder rrr = ReadRowsRequest.newBuilder()
.setTableNameBytes(ByteStringer.wrap(lastTableBytes))
.setFilter(filter)
- .setRowRange(range);
+ .setRows(rowSet);
List<Row> rows;
try {
@@ -372,19 +385,14 @@ public class GoogleBigtableClient extends com.yahoo.ycsb.DB {
try {
if (clientSideBuffering) {
- asyncExecutor.mutateRowAsync(rowMutation.build());
+ bulkMutation.add(rowMutation.build());
} else {
client.mutateRow(rowMutation.build());
}
return Status.OK;
- } catch (ServiceException e) {
+ } catch (RuntimeException e) {
System.err.println("Failed to insert key: " + key + " " + e.getMessage());
return Status.ERROR;
- } catch (InterruptedException e) {
- System.err.println("Interrupted while inserting key: " + key + " "
- + e.getMessage());
- Thread.currentThread().interrupt();
- return Status.ERROR; // never get here, but lets make the compiler happy
}
}
@@ -410,19 +418,14 @@ public class GoogleBigtableClient extends com.yahoo.ycsb.DB {
try {
if (clientSideBuffering) {
- asyncExecutor.mutateRowAsync(rowMutation.build());
+ bulkMutation.add(rowMutation.build());
} else {
client.mutateRow(rowMutation.build());
}
return Status.OK;
- } catch (ServiceException e) {
+ } catch (RuntimeException e) {
System.err.println("Failed to delete key: " + key + " " + e.getMessage());
return Status.ERROR;
- } catch (InterruptedException e) {
- System.err.println("Interrupted while delete key: " + key + " "
- + e.getMessage());
- Thread.currentThread().interrupt();
- return Status.ERROR; // never get here, but lets make the compiler happy
}
}
@@ -434,11 +437,18 @@ public class GoogleBigtableClient extends com.yahoo.ycsb.DB {
private void setTable(final String table) {
if (!lastTable.equals(table)) {
lastTable = table;
- lastTableBytes = options
- .getClusterName()
- .toTableName(table)
+ BigtableTableName tableName = options
+ .getInstanceName()
+ .toTableName(table);
+ lastTableBytes = tableName
.toString()
.getBytes();
+ synchronized(this) {
+ if (bulkMutation != null) {
+ bulkMutation.flush();
+ }
+ bulkMutation = session.createBulkMutation(tableName, asyncExecutor);
+ }
}
}
diff --git a/pom.xml b/pom.xml
index 96ea2ceb..9910871e 100644
--- a/pom.xml
+++ b/pom.xml
@@ -77,7 +77,7 @@ LICENSE file.
<cassandra.cql.version>3.0.0</cassandra.cql.version>
<geode.version>1.2.0</geode.version>
<azuredocumentdb.version>1.8.1</azuredocumentdb.version>
- <googlebigtable.version>0.2.3</googlebigtable.version>
+ <googlebigtable.version>0.9.7</googlebigtable.version>
<infinispan.version>7.2.2.Final</infinispan.version>
<kudu.version>1.1.0</kudu.version>
<openjpa.jdbc.version>2.1.1</openjpa.jdbc.version>
--
GitLab