MapR-ES Frequently Asked Questions

Document created by maprcommunity Employee on Jan 27, 2017Last modified by aalvarez on Jun 7, 2017
Version 7Show Document
  • View in full screen mode

 

Get Started

What overhead does MapR Streams introduce for messages? 

Editing MapR streams parameter 

MapR Streams CLI Tutorial

MapR Streams is built on MapR-DB tables. Tell me more. 

If I consider MapR data nodes as MapR Streams brokers (using Kafka terminology), how then does MapR Streams use mapr cluster resources? 

How fast is MapR Streams? 

 

Kafka

How compatible is MapR-Streams with Kafka? 

MapR-Streams supports a very large number of topics in comparison to Kafka. How does it do this? 

Using KafkaConsumer to subscribe topics with a Pattern 

I’m writing code in MapR Streams to write to a topic. It seems like my messages aren’t being sent as quickly as they are sent with Kafka. Each send seems to take 3 seconds. In Kafka, it’s super fast. Why? 

Does MapR Streams use ZooKeeper like Kafka does? 

Do any Kafka properties differ from MapR Streams properties? 

Will MapR Streams support Kafka 0.10 and Kafka Streams for processing? 

What is the Kafka REST Proxy for MapR Streams? 

Kafka Connect HDFS Sink and MapR FS 

Kafka listens on TCP port 9092. What port does MapR Streams listen on? I want to point a Kafka client at a MapR Streams broker. 

 

 

Performance

I’m writing code in MapR Streams to write to a topic. It seems like my messages aren’t being sent as quickly as they are sent with Kafka. Each send seems to take 3 seconds. In Kafka, it’s super fast. Why? 

What latency does MapR Streams provide for message delivery? 

It’s said that MapR Streams can handle an unlimited number of topics per stream. What are the scalability implications of this? 

How does MapR Streams handle committing cursors? Are there scalability concerns? 

How fast is MapR Streams? 

 

Configuration

Messages in MapR stream expiration (default TTL passed), can messages be recovered? 

Changing the retention times in MapR streams 

Can MapR Streams messages be stored infinitely? 

Since MapR Streams uses the MapR replication gateway can it support complex replication topologies like MapR-DB? 

How can I send messages synchronously? I don’t want to send the next message until the previous message has been received. 

Is there guts output from MapR Streams? 

If new consumers are added to an active consumer group, partitions are reassigned. How do I prevent duplicate messages? 

What recommendations do you have for achieving maximum throughput for a MapR Streams stream? 

Can I subscribe to a topic and seek() and get information about offsets without reading or polling for messages? 

Kafka listens on TCP port 9092. What port does MapR Streams listen on? I want to point a Kafka client at a MapR Streams broker. 

 

Integration

Does MapR Streams work with Logstash? | How to use logstash with MapR Streams 

MapR Streams via Zookeeper 

MapR Streams via PySpark 

Querying MapR streams using Hbase 

Does Streamsets work with MapR-Streams? 

Confluent's schema-registry 

Writing back into MapR streams using Spark (Java) 

How compatible is MapR-Streams with Kafka? 

Will MapR Streams support Kafka 0.10 and Kafka Streams for processing? 

Does MapR Streams work with Sqoop2? 

Does MapR Streams work with Spark? 

 

2 people found this helpful

Attachments

    Outcomes