This Glossary explains all the terms you’ll see in Beeks documentation. To report any omissions, please contact documentation@beeksgroup.com.

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

A

  • Access IP — The Access IP is the IP address a tenant uses to connect to a compute device.
  • ACD (Advanced Configurable Decoder) — The Advanced Configurable Decoder (ACD) is a Beeks Analytics decoder that clients can customise to meet their own requirement.
  • Agent — An Agent Point is an end point in VMX-Analysis, responsible for translating business data events from probes, and converting and mapping them to the Attributes for the relevant Flow. 
  • Agent Event — An Agent Event is a subset of important Attributes that Beeks Analytics passes from the message to persistent storage or to VMX-Analysis.
  • Aggregation — An Aggregation is a set of specific calculations applied to data or previously calculated statistics by an Aggregator. For example, an aggregation might calculate a single value, such as the average time (in seconds) it takes for RFQs to receive a price within a 10-second window.
  • Aggregator — An Aggregator defines how data is “sliced and diced“ to produce statistics - or stats - that are then displayed in the dashboards in VMX-Explorer.
  • Alert — An Alert (in Beeks Analytics terms) is a message generated by VMX-Analysis or VMX-Explorer when a time series (e.g. latency) breaches a particular predefined threshold, or when certain predetermined triggers are received (e.g. market data gaps).
  • Alt ID — An Alternative ID - or Alt ID - is another ID that we use to correlate Agent Events when a common Correlation ID is not available.
  • Anomaly — In data analysis, an Anomaly is some data that differs from the expected pattern of data. The most common type of anomaly is a market data gap.
  • Apache Arrow Apache Arrow is a language-agnostic software framework that offers high-speed in-memory data processing in data analytics applications that process columnar data.
  • App-to-wire latency App-to-wire latency is the time between the SendingTime timestamp in a message and the time it is observed at the Visibility Point. It is also known as the Wire time. 
  • Appliance — An Appliance is a machine with Beeks hardware, running a Beeks image.
  • Artificial Intelligence (AI) Artificial Intelligence (AI) is the simulation of intelligence in machines, enabling them to perform tasks that typically require human cognition.
  • Asset — The dictionary definition of an Asset is a ‘useful or valuable thing’. For IT infrastructure (and most commonly when Beeks use the term), this often means tangible assets such as computer hardware, network switches, etc. The Beeks Portal allows customers to manage their IT assets.
  • Association — An Association represents a group of related Items. Whereas Correlation concerns grouping multiple Agent Events into a single Item (aka Business Object), Association groups multiple related Items together.

Back to top

B

  • Bare Metal Bare metal is a computer system without a base operating system (OS) or installed applications.
  • Beeks Analytics Beeks Analytics is a solution that records and analyses latencies at network and application level, delivering exceptional real-time insight into performance of your trading environment.
  • Beeks Portal — The Beeks Portal is a secure, web-based user interface that enables users to order, oversee, and manage compute in their Beeks environment.
  • Berkeley Packet Filter (BPF) — A Berkeley Packet Filter (BPF) is a filter that you can use in a packet capture tool to retrieve only the subset of network packets that you are interested in.
  • Bill of Materials (BOM) — A Bill of Materials is a document that lists all the devices, cables, and sundries for a tenant.
  • Blade — A Blade (also known as an expansion module) is hardware installed into a chassis.
  • Branch node — A Branch node is a level in an Aggregation.
  • Broker — Specifically at Beeks, a Broker is a Beeks tenant that uses Beeks' cloud-based global network to achieve low latency execution and exchange connectivity.
  • Build Order — A Build Order is a request from the tenant to either provision or order compute.
  • Business object — A Business object is an instruction or entity such as an order or a price.

Back to top

C

  • Cancel-to-confirm Cancel-to-confirm latency is the time period between submission of a cancel request by a trader and the trader receiving confirmation of that cancel request back from the trading system.
  • Capture Attribute — A Capture Attribute is a name and a data type, which Agents use to store data about a business object. For example, the quantity and price of an order.
  • Channel — A Channel (in a Market Data context) is a multicast group and port tuple. VMX-Capture probes are configured to listen to channels to get feed data.
  • Chassis — A Chassis is a physical metal frame that allows you to store multiple servers and other hardware in a stack to save physical space in the data centre.
  • Child tenant — Proximity Cloud and Exchange Cloud allow Beeks tenants (“Parents“) to offer cloud services to their own customers or sub-organisations (“Children“). A Child tenant is a customer of the Parent tenant.
  • Client tenant — Proximity Cloud and Exchange Cloud allow Beeks tenants (“Parents“) to offer cloud services to their own customers or sub-organisations. A Client tenant is a sub-organisation within the Parent tenant that has its own dedicated Beeks Portal.
  • Colo (Co-location) — In financial markets, Colos are data centres with high-speed access to trading venues, which provide servers or compute services in the data center to its clients to enable low latency trading.
  • Compute — In a data centre, Compute is the processing power and memory required to run applications on a server.
  • Connectivity Hint — When a business object is broadcast to multiple end points, a Connectivity Hint is a piece of data injected into the business object by an Agent to indicate which branch the Agent Event is on.
  • Conversation — A network conversation is traffic between two specific endpoints.
  • Copula-based Outlier Detection (COPOD) — Unsupervised Copula-based Outlier Detection (COPOD) is technique for detecting data outliers that appear in the tails of a distribution.
  • Core Data Feed — Beeks Core Data Feed (CDF) provides a live streaming feed of data from VMX-Analysis and VMX-Capture.
  • Correlation Correlation is the act of grouping together all Agent Events for a given business object so that VMX-Analysis can track it as it moves through the network and/or the application.
  • Correlation Beacon — A Correlation Beacon is an identifier that is transmitted to Beeks Analytics that we can use to correlate items which otherwise would not have enough common information to allow them to be correlated.
  • Correlation ID — A Correlation ID (“CID“) is a special Capture Attribute which can be used to correlate multiple Agent Events together to form an Item.

Back to top

D

  • Darts DARTS is an open-source Python library that supports time series forecasting and anomaly detection.
  • Data Centre — A Data Centre is a physical facility that houses computer systems and associated components, such as networking and storage systems.
  • Decoder — A Decoder is a tool that transforms data into Beeks format, so that it can be analysed by Beeks Analytics.
  • Dedicated Server — A Dedicated Server (DS) is a physical server that is used exclusively by a single Beeks client.

Back to top

E

  • Empirical Cumulative-distribution-based Outlier Detection (ECOD) — Unsupervised Empirical Cumulative-distribution-based Outlier Detection (ECOD) is technique for detecting data outliers that appear in the tails of a distribution.
  • Exchange Cloud Exchange Cloud is a Beeks product that provides a multi-homed, fully configured and pre-installed physical trading environment that enables Exchanges to offer private cloud solutions as a service to their customers.
  • External Group — An External Group is a set of connections to external counterparties or datafeeds that are monitored together to provide metrics about a given type of traffic.

Back to top

F

  • Feeds — A Feed is a data stream that provides price and other information to enable traders to choose to buy or sell securities and commodities.
  • Fill ratio — The Fill ratio (or Fill rate) is the percentage of an order that is successfully executed or “filled“.
  • FIX Protocol FIX (Financial Information eXchange protocol) is a protocol that is used by financial participants to exchange market data and trading information.
  • Flow — A Flow is the journey of a business object type, such as an order or trade, through the monitored system.
  • Forecasting — In Beeks, forecasting is the prediction of future infrastructure and trading performance based on historic data and modelling.

Back to top

G

  • Gap Detection — A Gap is where one or more packets in a sequenced data feed that don’t reach their destination. Gap detection is real-time monitoring of when these gaps occur.
  • GPU (Graphical Processing Unit) GPU is a technology that was originally designed to support accelerated rendering of 3D graphics but which is now used for other computing purposes that require high-speed processing because of its extraordinary amount of computational capability.

Back to top

H

  • Hit Ratio Hit ratio refers to the proportion of trade requests or orders that are successfully executed relative to the total number of trade requests or orders submitted. It is a measure of how often a trading strategy or trading system achieves its intended trades.
  • Hypervisor — A hypervisor is computer software, firmware, or hardware that creates and runs virtual machines.

Back to top

I

  • Internal Group — An Internal Group is one or both ends of connections to internal compute such as a Dedicated Server that are monitored together to provide metrics about these types of traffic.
  • Interval — An Interval is a set of information about a time period between two Agent Events.
  • IP (Internet Protocol) — In computer networking, the Internet Protocol (IP) is a protocol that enables data transfer between two entities.
  • Item — An Item represents all the Agent Events that relate to a single business object.
  • Item History — The Item History lists the captured data about a business object, the probe ID, and time the probe saw the item.
  • Item persistence Item persistence is where Beeks Analytics writes item information to a datastore.

Back to top

J

  • Jitter Jitter is when there is a time delay in the sending of data packets over a network.
  • JMX JMX (Java Management Extensions) is a Java technology that provides tools for managing and monitoring applications, system objects, devices, and service-oriented networks.
  • Jupyter Notebook Jupiter Notebook is a tool that allows users to enter code into “notebooks“ and view the corresponding outputs such as visualisations, text, and media.

Back to top

K

  • Kafka Kafka is an open-source distributed event streaming platform that’s developed by the Apache Software Foundation.

Back to top

L

  • Last Look — In financial markets, last look is a practice primarily used in the foreign exchange (FX) and over-the-counter (OTC) markets. It grants liquidity providers, such as banks or market makers, a brief window of time to accept or reject a trade request from a counterparty after the trade request has been made but before it is executed.
  • Latency Latency is the time it takes for a message to travel from one place to another in the trading environment.
  • Leaf node — A Leaf node is the lowest and final level in an Aggregation.
  • Log (OpenTelemetry) — In the context of OpenTelemetry, a Log is a timestamped message emitted by services or other components. These are not necessarily related to a specific business transaction.
  • Logical Cross Connect — A Logical Cross Connect is a direct and secure low latency fibre connection within shared data centre facilities that can significantly reduce latency, improve system performance, and minimise networking costs.
  • Low latency Low latency describes a computer network that is optimized to process a very high volume of data messages with minimal delay (latency).

Back to top

M

  • Market — Financial Markets are physical or digital locations in which traders buy and sell assets such as stocks, bonds, commodities, and derivatives. For example, NYSE: ARCA, NYSE Equities, American Equities, Chicago Equities.
  • Market Data Market data is real-time or historical information about a financial market or instrument. It typically provides information such as price, bid/ask quotes, and market volume.
  • Market Edge Intelligence — Beeks Market Edge Intelligence provides a modular low-latency high-volume AL/ML platform for analyzing front-office network, trading and market data to support real-time decision making, anomaly detection, and increased trading profitability.
  • Match Attribute Match Attributes are Capture Attributes that VMX-Analysis will group together to provide a unique signature for a business object so that it can identify the messages that belong to an Item.
  • mdPlay — In Beeks Analytics, mdPlay is a standalone replay-only version of the main Beek Analytics VMX-Capture component.
  • Message Collector — A Message Collector is a program running in a probe that captures all messages and writes it to a downstream store such as InfluxDB.
  • Microburst — A microburst describes an event when a large amount of burst data is received in milliseconds.
  • Multicast — A multicast is a transmission of data from a single source to multiple recipients.
  • Multivariate analysis Multivariate analysis is a category of data analytics that looks at data containing more than two variables.

Back to top

N

  • NapaTech Programming Language (NTPL) NapaTech Programming Language (NTPL) is a simple plain text scripting language that can be used for advanced configuration in Beeks Analytics.
  • Network Timing Protocol (NTP) Network Timing Protocol (NTP) is the standard timing technology used on computer networks.
  • Nodepath — A Nodepath is a way of defining the levels in an Aggregation.

Back to top

O

  • Open System Interconnection (OSI) OSI is a network model that was developed by ISO in 1978.
  • OpenTelemetry OpenTelemetry is an open source observability framework that offers protocols and tools for collecting and routing telemetry data.
  • Order-to-acknowledgement Order-to-acknowledgement is the time period between a trader submitting an order and the order being acknowledged by the trading system.
  • Order-to-first-fill Order-to-first-fill is the time period between a trader submitting an order and when the first portion of that order is executed or “filled“.
  • Order-to-tick Order-to-tick is a measurement specifically related to a Matching Engine. It refers to the time taken between receiving an order from a specific market participant and sending out a market data update to inform all market participants of the new order.

Back to top

P

  • P3 — The P3 process is part of the VMX-Capture architecture. It is called P3 because it performs three functions: pre filtering, pre correlation, and pre aggregation.
  • Packet — A Packet is a basic unit of data that's transferred over a computer network.
  • Packet Broker — A Packet Broker is a network device which can receive, buffer, filter and distribute packets to various monitoring and security tools.
  • Packet Capture probe — A Packet Capture probe captures packets and writes them to disk to provide a store of the captured data, which users can then consult if needed for diagnostics or evidence.
  • Packet Capture Repository — A Packet Capture Repository is a data store of packets that have been picked up by probes in Beeks Analytics.
  • Packet Multiplexer (PMUX) — A Packet Multiplexer (PMUX) is a mechanism by which a packet stream is filtered in order to split traffic for capture to disk or for decoding to make analytics possible. 
  • Parent tenant — A Parent tenant is a Proximity Cloud or Exchange Cloud tenant that provides Beeks cloud services to their clients or sub organisations.
  • Parquet — Apache Parquet is an open-source column-oriented data storage format that supports large data sets because it’s optimised for disk I/O and can achieve high compression ratios with columnar data.
  • Port mirroring Port mirroring is a technology that sends a copy of the packets seen on one switch port (or an entire VLAN) to a network monitoring connection on another switch port.
  • Pre-aggregated Stats — A Pre-aggregated stat is a set of data that has been aggregated in VMX-Capture and the result sent to VMX-Analysis.
  • Precision Time Protocol (PTP) PTP, or Precision Time Protocol, is a network-based time synchronization standard, that is designed to provide synchronisation at a nanosecond or even picosecond level.
  • Prism — See VMX-Prism
  • Probe — A Probe is an ultra-fast, lightweight capture mechanism within VMX-Capture which decodes and/or analyses traffic, delivering business objects and statistics to one or more Agent Points configured in Beeks Analytics.
  • Proximity Cloud Proximity Cloud is a Beeks' product that offers a private environment in a fully configured, pre-built rack that can be deployed to any location in the world and that is designed for trading in capital markets.
  • PyOD PyOD (Python Outlier Detection) is an easy-to-use Python library for detecting anomalies in multivariate data.

Back to top

Q

  • Query — A Query is request for data that is sent to a data source.

Back to top

R

  • Rack — A Rack is a frame for stacking hardware such as servers and switches. The Beeks Portal includes a view that enables you to see what’s in your racks and how much capacity is remaining.
  • Reject rate — The Reject rate is the percentage of orders that are not accepted - or “rejected” - by the trading system or counterparty compared to the total number of order submitted.
  • Root cause analysis Root cause analysis is the analysis of large sets of data around the time of a known problem, in order to identify its cause.
  • Root node — A Root node is the first level in an Aggregation.
  • Roundtrip latency Round trip latency (or Request/Response latency) is the time period between a message being observed at an Agent Point and the message response being observed at the same Agent Point.

Back to top

S

  • Scalable Unsupervised Outlier Detection (SUOD) Scalable Unsupervised Outlier Detection (SUOD) is an accelerated framework for training and prediction in ML models that perform outlier detection.
  • Searchable Attribute — A Searchable Attribute is a specially designated Capture Attribute that you can use to filter items in a list. 
  • Sequence Sequence is a group of messages for a given business object, arranged into the order in which they were sent originally.
  • Session — Within Beeks Analytics, a Session is a field that is used to record the FIX CompIDs of the FIX session. For non-FIX traffic, this field will be populated with IP addresses.
  • Side — A Side is a feed within a feed. Most trading venues produce their raw feeds as a pair of UDP multicast feeds. We refer to these two UDP multicast feeds as the A-side and B-side.
  • Similarity analysis Similarity analysis measures how similar two objects are to each other; it is essential to recommendation systems, clustering, and anomaly detection.
  • Simple Binary Encoding (SBE) Simple Binary Encoding is a binary representation for encoding/decoding messages to support low-latency streaming.
  • Span — A Span is an OpenTelemetry term. In Beeks Analytics, an Agent Event can be considered a Span.
  • Stack probe — A Stack probe is a probe that decodes of messages, timestamps, maps, detects anomalies, and generates pre-aggregated stats. It then sends the decoded message, anomalies, and pre-aggregated stats to the Agent service in VMX-Analysis and to a Core Data Feed.
  • Statistics Collector — A Statistics Collector (or stat collector) is a program in a stack probe that creates pre-aggregated statistics from the observed traffic, which it then writes to a downstream component such as VMX-Analysis.
  • stream2Cloud — Beeks stream2Cloud was a lite analytics tool delivering high-level network analytics, basic financial protocol analysis (market data gap detection) and capture, and network telemetry.

Back to top

T

  • Tapping — A network tap is a device that makes it possible to access network traffic, in order to monitor network events.
  • TCP (Transmission Control Protocol) — In computer networking, the Transmission Control Protocol (TCP) is a protocol that enables data transfer between two entities.
  • Tenant — A Tenant is a group of users that have been given access with specific privileges to a set of cloud-based software or hardware resources.
  • Tick — A Tick is a price movement in financial markets.
  • Tick-to-order Tick-to-order is the time period between a trading system receiving a market tick (price change) and it generating an order and sending the order to the exchange.
  • Tick-to-trade Tick-to-trade is the time it takes for a trade to be executed from the moment the decision to trade is made. It includes the time taken for the order to be placed, transmitted, matched, and finally executed on the exchange.
  • Timeseries — A Timeseries is a series of data points indexed (or listed or graphed) in time order. In Beeks Analytics, timeseries is usually associated with a given Aggregator cell, showing variation over time.
  • Topic — In messaging systems such as ActiveMQ and Kafka, a Topic is a category used to group messages of the same subject or that are used for a common purpose. Producers publish messages to a given Topic and Consumers then subscribe to the Topics they want to receive messages for.
  • Trace — A Trace is a record of all events that occurred for a specified Business object as it travelled through the trading architecture.

Back to top

U

  • UDP (User Datagram Protocol) — In computer networking, the User Datagram Protocol (TCP) is a protocol that enables data transfer between two entities.
  • Unicast Unicast is a one-to-one transmission from one point in the network to another point.
  • Univariate analysis Univariate analysis is a category of data analytics that looks at data with just one variable.

Back to top

V

  • Venue Venue is a digital platform or physical location in which financial trading occurs, for example, exchanges like NYSE, ICE, Nasdaq.
  • VEx (VMX Expression Language) VMX Expression Language (VEx) is a simple expression language used to configure Beeks Analytics.
  • Virtual Compute Virtual Compute is a virtual machine, running on a hypervisor.
  • Virtual Private Server — A Virtual Private Server (VPS) is an isolated, private environment on a physical server.
  • Visibility Point — A Visibility Point is a location in an application or network from which we want to capture data.
  • VLAN Tagging VLAN (Virtual Local Area Network) tagging is where network packets are each assigned a VLAN tag in the Ethernet frame.
  • VMX-AdminConsole VMX-AdminConsole is a user interface tool that enables you to configure and maintain Beeks Analytics Enterprise.
  • VMX-Analysis — In Beeks Analytics, VMX-Analysis is the layer that analyses the data from VMX-Capture to produce reports and views, and passes aggregated data and individual business items to VMX-Explorer for visualisation.
  • VMX-Capture — In Beeks Analytics, VMX-Capture is the layer responsible for data capture and decoding. 
  • VMX-Explorer — In Beeks Analytics, VMX-Explorer is a Grafana-based web user interface, and is where Beeks Analytics visualises the analytics provided by VMX-Analysis. VMX-Explorer is fed by the Beeks Analytics REST API.
  • VMX-Prism VMX-Prism is a Beeks Analytics packet capture system that runs on your appliance to support advanced querying, analysis, and decoding of packet captures.
  • VPS Cluster — A VPS Cluster is a collection of multiple VPS Hosts.
  • VPS Host — A VPS Host is a physical server, running a hypervisor, which hosts one or more Virtual Private Servers (virtual machines).

Back to top

W

  • Wiretime Wire time is a metric for measuring latency. It is the time that Beeks Analytics calculates between the SendingTime timestamp written within the message and the time it is observed at the Visibility Point.

Back to top

X

  • XMLRPC XMLRPC is a third party spec and a set of implementations that allow software running on disparate operating systems, running in different environments to make procedure calls over the Internet.

Back to top

Y

Back to top

Z

Back to top