In this paper, we propose an array architecture that supports in-memory computation based on a logic array first proposed in 1972 by Sheldon Akers.
Future work will include looking at how the data pipeline in the new approach will perform if the MapReduce framework were to be replaced by the Spark ecosystem which supports in-memory processing [21].
Spark supports in-memory computing, that enables it to query data much faster compared to diskbased engines such as Hadoop, and supports a rich set of other tools such as Spark SQL (for SQL and structured data processing), MLLib (for machine learning) and Spark Streaming.
Neo4j does not support graph sharding on different nodes, rather it supports in memory cache sharding [52, 82].
Spark is best suited for Adaptive-Miner because it has support for in-memory distributed computation.
For future work, there is still more work that can be conducted to optimize iiHadoop to support fast in-memory processing, and comparing it with other in-memory systems such as Spark and Naiad.
Without using any of the new features in the software, which include better support for in-memory computing, for example, queries should execute 25percentt faster on the same hardware.
Modern monitoring use cases however go beyond simple graphing and necessitate a much wider range of processing, this requires greater support for in memory storage than what stock rrdtool provides.
Easiest, perhaps, is the simple act of finding your local animal shelter and offering the gift of time (volunteering) or financial support donated in memory of the slain lion.
This effect finds some support in the memory literature as well: Untrained observers expect reminisced memories to be much lower in accuracy than consistently produced information, and even to be lower in accuracy than they actually are (Oeberst, 2012).
