Payotek
Data Without Borders
Aggregate Accelerate Analyze
We deliver a unified, self-service acceleration and data federation platform for business analysts, data scientists and data engineers in today’s multi and hybrid cloud world.
Absolute Advantage
SELF-SERVICE
HYBRID CLOUD
Payotek Hub provides a unified view across data lake, datastream, data warehouse, RDBMS sources, deployed in a hybrid cloud that spans across on-premises and multi-cloudenvironments.
SECURE WITH DATA
PRIVACY CONTROL
With Payotek’s Hub, you can mask all the columns of your tabular data (e.g. personal information) or anonymize bi-directionally. You can have end to end encryption with in-flight and at-restencryption to achieve data privacy in a hybrid environment.
ACCELERATED BUSINESS
INTELLIGENCE HYBRID CLOUD
Business analysts and citizen data scientists can achieve 30-80x faster performance with their interactive queries when using their
favorite BI tools with Ampool.
SQL QUERY FEDERATION A
CROSS DATA SILOS
Business analysts can join complex datasets across disparate data silos spanning across on-premises and cloud environments at a speed, not possible with today’s legacy data
federation vendors.
REAL-TIME AND
HISTORICAL DATA
With a single unified Ampool Hub and the connectors, you can handle both real-time data (e.g. real-time streams in Kafka broker) and historical data in data lake and data.
EXTENSIBLE TO MACHINE
LEARNING
Seamlessly adopt predictive insights with Machine Learning by enabling your existing business analysts and citizen data scientists’ access to disparate datasets from the same unified view.
Applications and Tools
Custom Apps (PP)
payotek
Payotek Proxy (PP)
Data transformation: Two-way hashing, masking etc, Query pushdown
MULTI-SOURCE CONNECTORS
Ampool supports standard connectors to variety of data sources -data lake (Cloudera Data Platform, Azure HDInsight, Amazon EMR), data warehouse (Snowflake, Amazon Redshift), real-time data stream (Confluent, Cloudera Data Flow) and relational database (Oracle, MySQL, SQL Server, PostgreSQL). Ampool also supports custom connectors, depending on the data source.
CLOUD NATIVE AND AUTO SCALING
Each Ampool is deployed as a set of docker containers, that are independently scalable and leverages RESTinterface for external interfaces and User Interface. Ampool supports service deployment using Kubernetesand Helm charts and enables auto-scaling with metricsbased triggers. Ampool can provide a consistent
experience across Amazon AKS, Microsoft EKS, Google Anthos, OpenShift, and VMware Tanzu.
SINGLE PANE ACROSS HYBRID CLOUD
The Ampool Hub provides a self-service unified single pane for administration, resource control and user management. With role-based access control, each user self-serve, see his/her data sources, run analytics on-demand. The interface remains unchanged as the customer goes through a hybrid and multi-cloud journey.
DATA FEDERATION
With Ampool’s Query/Data Federation and Transparent Caching, there is no need to pre-load data into the Ampool Analytics Engine, unlike traditional vendors. A single SQL query can operator on datasets across
multiple data sources. Ampool transparently caches (materializes) frequently used subset datasets. Queries are rewritten to access materialized datasets, if applicable.
PERFORMANCE
Ampool is up to 40x faster than Amazon EMR and up to 3x faster than Amazon Redshift on BI queries. When it comes to data loading, scalability, and concurrency, Ampool is 3-20x faster. Ampool is up to 80x faster than Hive on complex TPC-DS queries.
DATA MODELING LANGUAGE
Ampool maintains JSON-based specification of relationships between datasets across data sources. It supports primary key to foreign key relationship, uniqueness constraints, 1 to many constraints on fields
from multiple tables. This is extensible for preaggregation, building cubes, and ML model computations.
COST BASED OPTIMIZER
The Ampool Analytics Engine (AE) analyzes loaded datasets, maintains statistics and updates in the Ampool metadata store. It does join re-ordering,
filter, projections, and limit pushdowns.
DATA GOVERNANCE AND ANONYMIZATION
The Ampool Caching Proxy (ACP) enables one-way hash or bi-directional tokenization of sensitive data form data sources. Data sink exports templates to securely decrypt tokenized data..
ENTERPRISE SECURITY
AND PRIVACYR
The Ampool Hub integrates with Kerberos and can form a trusted relationship with an organization’s KDC. Ampool supports end to end encryption on
the wire as well at-rest encryption. Ampool also enables masking of columnar data such as social security number of all customers in a table before the data is shared for analytics. Ampool support role-based access control (RBAC) and supports personas such as administrators and end users.
COST CONTROL WITH SUSPEND AND RESUME
The Ampool Hub enables fast backup to encrypted cloud-based object store and suspension of computeclusters. It resumes by spinning up the compute
clusters and fast restoring from encrypted cloud object store. End users can schedule to not use cloud hosted cost-based optimizations for clusters with apre-defined schedule (5pm-9am
AI/ML FRAMWORKS
Ampool empowers data scientists to connect to frameworks such as Apache Spark, TensorFlow, PyTorch as