The Platform designed by Miras code named LeggoApp has its own proprietary distributed object file system, called ActorFS™, which was released in July of 2014 at the ICT Spring conference in Luxembourg and has achieved significant interest from industry and end users. ActorFS is a Parallel Object File System that competes with the best of class of file systems in the world and has bench marked itself to show its capabilities against the existing market leaders in a superior way.
ActorFS is disruptive in that it:
Is 24 times faster than Hadoop Map/Reduce Model currently in use by the industry while being fully compatible with Map/Reduce
Is 2.3 times faster than in-memory processing technologies
With Vector Processing Method adaptation is more than 10 times faster than conventional query engines
The significant advantages of ActorFS™ in terms of processing speed (50 times faster with Parallel read on small size objects that are less than 128MB, also faster for latency for large objects), size and structure of its data-blocks, its ability to address streaming data and faster parallel scan (sequential read) will position it to become the premier file system available in the market place. The business models of “Data as a Service” and “Platform as a Service” (PaaS) uniquely position the Company to offer the enterprise segment this set of capabilities and tools at price-performance ratios that are not available by any other competitor in the market place and cannot be offered /matched by existing technologies. ActorFS bench marks are provided in this brief as part of technology descriptions.
LeggoApp enables enterprises to use latest best technologies to harvest enormous amount of unstructured data and get access to on-demand query on all enterprise data sources. This capability of dealing with data provides real-time analytic and enables decisions based on data. The strong demand for trials for the system are a good indication of the demand in the market for LeggoApp and what it offers.
In LeggoApp data will be compressed to 1/10 of original size, compared to Microsoft SQL Server, which will reduce the costs of analytic platform (like conventional data warehouses) dramatically and also reduces the surveying time through huge amount of data by fully utilizing the underlying hardware.
LeggoApp is designed to be easy-to-install and easy-to-use. Software installation and connecting to data sources will not take more than a day and from the first day users can start digging in data through Full-text search and standard SQL.
LeggoApp was designed to minimize the switching costs for the enterprise clients and has achieved this capability by offering a simple web based interface. LeggoApp is Hadoop HDFS, and POSIX compatible thus all Hadoop users will use LeggoApp just as they do Hadoop, seamlessly. Legacy Linux Applications such as web server and media and file servers will also interface with the system seamlessly through POSIX interface that is common for such applications.
No need to configure a solution to be fit on enterprise network
No need to develop an in-house solution
Using just systems that are necessary for enterprise needs
Connect to Data sources, gather it and make it available for analytic
Don’t worry about size, format and structure of data or bandwidth of transfer
Don’t think about paying expensive price
Columnar execution engine
times faster than Microsoft SQL Server
Supports data locality that reduces network traffic when LeggoApp is co-located with the data source
Seamlessly compress-decompress data on query
High performance and low latency analytic
Reduce the cost of infrastructure dramatically
Correlate data from different data sources
Join Oracle database tables with Hadoop log files
Connecting immediately to new data sources and start query data
ndex-less and cube-less data warehouse
High-Performance query on all data fields without requiring defining index or cube
Concurrent query support by optimized query queueing processing
Ready for cloud applications
Data replication for availability in data loss scenarios
Process delegation when one processing node becomes unavailable
Distributed query processing elastically scalable to 1000 nodes
Data scalability separated from process scalability to maximize flexibility in adding resources
Interact with data in standard SQL language
Standard ODBC/JDBC interface
unstructured data loader
Loads data into indexing engine (from database, file, network packets and web)
Full-text engine which helps to search through your data
Visualization engine and dashboard
structured data loader
Loads data for SQL analytic
structured storage engine
Stores data for SQL analytic in compressed form
structured data source
A connector when it’s necessary to not moving data