首页
学习
活动
专区
工具
TVP
发布
社区首页 >问答首页 >对Cassandra 4节点群集进行基准测试的配置是什么

对Cassandra 4节点群集进行基准测试的配置是什么
EN

Stack Overflow用户
提问于 2018-05-30 03:02:28
回答 1查看 182关注 0票数 -1

我在VMWare ESXi 6.7中运行4个节点的Cassandra集群,并使用ycsb测试4个节点的Cassandra集群。结果根本不像预期的那样,我只是在4节点集群的情况下获得了20000次操作/秒。可以告诉我基准测试的配置是什么吗?下面是我的配置文件。

代码语言:javascript
复制
cluster_name: 'test-cluster'
num_tokens: 256
hinted_handoff_enabled: true
max_hint_window_in_ms: 10800000 # 3 hours
hinted_handoff_throttle_in_kb: 1024
max_hints_delivery_threads: 2
hints_directory: /mnt/hints
hints_flush_period_in_ms: 10000
max_hints_file_size_in_mb: 128
batchlog_replay_throttle_in_kb: 1024
authenticator: AllowAllAuthenticator
authorizer: AllowAllAuthorizer
role_manager: CassandraRoleManager
roles_validity_in_ms: 2000
permissions_validity_in_ms: 2000
partitioner: org.apache.cassandra.dht.Murmur3Partitioner
data_file_directories:
    - /mnt/cassandra-data
commitlog_directory: /mnt/commitlog
disk_failure_policy: stop
commit_failure_policy: stop
key_cache_size_in_mb: 0
key_cache_save_period: 14400
row_cache_size_in_mb: 0
row_cache_save_period: 0
counter_cache_size_in_mb:
counter_cache_save_period: 7200
saved_caches_directory: /mnt/saved-cache
commitlog_sync: periodic
commitlog_sync_period_in_ms: 100
commitlog_segment_size_in_mb: 32
seed_provider:
    # Addresses of hosts that are deemed contact points. 
    # Cassandra nodes use this list of hosts to find each other and learn
    # the topology of the ring.  You must change this if you are running
    # multiple nodes!
    - class_name: org.apache.cassandra.locator.SimpleSeedProvider
      parameters:
          # seeds is actually a comma-delimited list of addresses.
          # Ex: "<ip1>,<ip2>,<ip3>"
          - seeds: "192.168.50.235,192.168.50.254"
concurrent_reads: 64
concurrent_writes: 32
concurrent_counter_writes: 32
concurrent_materialized_view_writes: 32
memtable_allocation_type: heap_buffers
memtable_flush_writers: 4
index_summary_capacity_in_mb:
index_summary_resize_interval_in_minutes: 60
trickle_fsync: true
trickle_fsync_interval_in_kb: 256
storage_port: 7000
ssl_storage_port: 7001
listen_address: 192.168.50.235
start_native_transport: true
native_transport_port: 9042
start_rpc: false
rpc_address: localhost
rpc_port: 9160
rpc_keepalive: true
rpc_server_type: sync
thrift_framed_transport_size_in_mb: 15
incremental_backups: false
snapshot_before_compaction: false
auto_snapshot: true
tombstone_warn_threshold: 1000
tombstone_failure_threshold: 100000
column_index_size_in_kb: 64
batch_size_warn_threshold_in_kb: 5
batch_size_fail_threshold_in_kb: 50
unlogged_batch_across_partitions_warn_threshold: 10
compaction_throughput_mb_per_sec: 0
compaction_large_partition_warning_threshold_mb: 100
sstable_preemptive_open_interval_in_mb: 50
read_request_timeout_in_ms: 5000
range_request_timeout_in_ms: 10000
write_request_timeout_in_ms: 2000
counter_write_request_timeout_in_ms: 5000
cas_contention_timeout_in_ms: 1000
truncate_request_timeout_in_ms: 60000
request_timeout_in_ms: 10000
cross_node_timeout: false
endpoint_snitch: GossipingPropertyFileSnitch
dynamic_snitch_update_interval_in_ms: 100 
dynamic_snitch_reset_interval_in_ms: 600000
dynamic_snitch_badness_threshold: 0.1
request_scheduler: org.apache.cassandra.scheduler.NoScheduler
server_encryption_options:
    internode_encryption: none
    keystore: conf/.keystore
    keystore_password: cassandra
    truststore: conf/.truststore
    truststore_password: cassandra
client_encryption_options:
    enabled: false
    optional: false
    keystore: conf/.keystore
    keystore_password: cassandra

internode_compression: all
inter_dc_tcp_nodelay: false
tracetype_query_ttl: 86400
tracetype_repair_ttl: 604800
gc_warn_threshold_in_ms: 1000
enable_user_defined_functions: false
enable_scripted_user_defined_functions: false
enable_materialized_views: true
windows_timer_interval: 1
EN

回答 1

Stack Overflow用户

发布于 2018-05-31 05:51:28

这实际上是一个具有单个磁盘的节点,在cassandra的4个实例之间存在竞争。确保每个实例都有自己的磁盘(任何数据库的运行速度都不能超过其磁盘)。如果想在一台主机上运行4个cassandra实例,可以使用ccm,或者只有4个不同的实例绑定到自己的接口,而不是使用虚拟化来降低速度。

ycsb在对cassandra进行基准测试方面相当糟糕,可以尝试使用cassandra附带的cassandra-stress。当推送超过100k时,您需要开始使用多个客户端(独立的系统),这也限制了单个客户端可以做什么。

票数 0
EN
页面原文内容由Stack Overflow提供。腾讯云小微IT领域专用引擎提供翻译支持
原文链接:

https://stackoverflow.com/questions/50591234

复制
相关文章

相似问题

领券
问题归档专栏文章快讯文章归档关键词归档开发者手册归档开发者手册 Section 归档