Progent's Email Guard uses the services and infrastructure of leading data security companies to provide centralized management and comprehensive protection for all your inbound and outbound email. The hybrid architecture of Email Guard combines a Cloud Protection Layer with an on-premises security gateway appliance to deliver world-class defense against spam, viruses, spoofing, phishing, spyware, Denial of Service (DoS) attacks, Directory Harvest Attacks (DHAs), and other email-borne threats.

Email Guard's Cloud Protection Layer acts as a first line of defense and keeps the vast majority of threats from reaching your security perimeter. This reduces your vulnerability to external attacks and conserves system resources. Email Guard's on-premises security gateway provides a deeper layer of inspection for inbound email. For outbound email, the local security gateway offers anti-virus and anti-spam filtering, policy-based Data Loss Prevention (DLP), and encryption. The local email security gateway can also manage your Exchange Server deployment to monitor and protect internal email traffic that never reaches your security gateway.

Email Guard works with any email platform and offers enhanced features for Exchange Server.

The architecture of Progent's Email Guard solution includes four main components that stand between the Internet and the email server at your data center:

  • Update Center
  • Cloud Protection Layer
  • Email Guard Security Gateway Appliance
  • Message Center
Email Guard's Architecture

Email Guard's hybrid architecture provides comprehensive protection for inbound and outbound email traffic

Update Center
The Update Center is a technology center staffed 24x7x365 by security experts who gather emails, URLs, binaries and other data collected from thousands of collection points worldwide, aggregate and analyze this information, and develop defenses, rules and signatures to update Email Guard's defense mechanism in order to provide zero-hour protection against the latest email-borne threats. Advanced filtering techniques and heuristic rules analysis used by the Update Center include reputation analysis, intent analysis, and predictive sender profiling. The result is a spam and virus filtering solution that blocks a very high percentage of threats yet delivers a very low number of false positives.

Cloud Protection Layer
The Cloud Protection Layer, which is kept current by the Update Center, provides spam and virus prefiltering using advanced technologies that include anti-fraud intelligence and blocking polymorphic viruses that can change their signatures to escape detection. Made up of multiple hardened datacenters worldwide, the Cloud Protection Layer has the capacity to handle Denial of Service attacks as well as email surges that occur during specific periods of the day. Because this cloud-based service stands between the Internet and the Email Guard security gateway, the majority of malicious email can be eliminated before it reaches your security perimeter. To help with disaster recovery and email continuity, the Cloud Protection Layer offers spooling. If your on-premises data center is disrupted, the spooling feature can store all inbound email in the cloud for up to 96 hours or direct it to an alternate destination while you repair the problem.

Email Guard Security Gateway Hardware Appliance
The Email Guard Security Gateway is a physical rack-mountable appliance that provides comprehensive protection for inbound, outbound and internal email. The gateway is automatically updated in real time by the Update Center without requiring management intervention. One of the gateway's deep inspection features is a multi-pass OCR engine that protects against image spam, where text is embedded within a graphic in order to hide content from spam filters. Another feature for filtering inbound email is link protection, which automatically evaluates and rewrites fraudulent URLs contained in a message. A plug-in agent for Microsoft Exchange Server offers virus protection for internal email traffic that does not access the gateway, for example emails from Outlook Web Access.

The Email Guard Security Gateway offers Data Loss Prevention (DLP) by allowing administrators to enforce content policies for outbound email in order to keep sensitive corporate information from being inadvertently leaked. The gateway can also be configured to block invalid bounce messages, which are outbound non-delivery reports to forged email addresses. In addition, the gateway offers a variety of encryption features for outbound email and supports inbound and outbound rate control. The Email Guard Security Gateway is available in several models to match your capacity requirements and security needs and can be clustered for higher performance and availability.

Message Center
The Message Center is a cloud-based service for encrypting outbound email. Emails that match policy are sent securely from the Email Guard security gateway to the Message Center. At the Message Center, the sender's email is encrypted using Advanced Encryption Service with a 256-bit cipher (AES-256) and stored for retrieval. The Message Centers sends the recipient a notification that includes a link to visit in order to retrieve the message. The link requires a web browser using HTTPS. The recipient selects a password and the Message Center delivers the encrypted message.

Email Guard's Outbound Encryption

The recipient only needs to establish a password for the first encrypted email received from the message center. Any replies are also sent via the Message Center to maintain security. Encryption key management is handled automatically by the Message Center, and the keys used to encrypt the data are stored in separate areas to provide physical security. Email traffic between the Email Guard Gateway and the Message Center and between the Message Center and the recipient are encrypted via SMTP over TLS.

In addition to policy-based encryption, Email Guard also provides an add-in for Microsoft Exchange Server that allows users to specify encryption from within Outlook.

Management Features
The Email Guard security gateway is integrated with a web-based management portal that provides a simple, centralized interface for viewing and managing all devices, services, policies and configurations. The cloud management tool offers global, domain-level, and individual user policy control. User-level policy management includes block lists, allow lists, quarantining, scoring thresholds, tagging and Bayesian analysis. At the domain level, you can manage policy for inbound and outbound quarantines, outbound encryption, plus block and allow lists based on IP address or sender/recipient email address or domain. Other management features include reports, graphs, statistics; recipient validation via LDAP; support for multiple domains; and delegated domain administration and help desk role.

Contact Progent for More Information about Email Guard Security Gateway Services
If you are interested in using Progent's Email Guard security gateway or need help with any aspect of email infrastructure or security, call 800-993-9400 or visit Contact Progent.

Glossary of Terms Related to Spam Filtering Technology

Term Definition

AES-256

The Advanced Encryption Standard (AES) is a cybersecurity specification adopted by the National Institute of Standards and Technology (NIST). AES-256 offers the highest level of encryption available within the AES family.

Allow List

Also called a whitelist, contains email addresses or domain names that are exempted from being blocked by a spam filter.

Anti-fraud Intelligence

Linguistic analysis to protect against fraudulent emails, which can take the form of spear phishing, spoofing, fee frauds or bogus offers of help.

Bayesian Analysis

A linguistic algorithm that develops profiles of the words and phrases used in legitimate email collected from specific users or organizations and determines whether new email from these sources fit their linguistic profile. Emails that are out of character have a higher likelihood of being spam.

Block List

Also called a blacklist, contains email addresses or domain names that are to be blocked by a spam filter.

Content Policies

Content policies allow administrators to define email filtering based on any combination of subject, headers, body, attachments, sender, or recipient. For example, specified keywords, categories or character sets can be automatically detected and blocked. Based on content policy, administrators can configure their email security gateway to block, allow or quarantine inbound or outbound email and to encrypt outbound email. Careful management of content policies can keep an organization from being blacklisted as a source of spam and other malware.

Data Loss Prevention (DLP)

Also called Data Leakage Prevention. Allows administrators to establish and enforce policies that block or encrypt outbound email that could release sensitive or valuable information. DLP features included in Email Guard can also help organizations avoid reputation loss and stay in compliance with industry security standards and government regulations such as HIPAA and Sarbanes-Oxley.

Delegated Administration

With Email Guard, control of domain-level settings can be assigned to the Domain Admin role, which has the most permissions, the Helpdesk role, with fewer permissions, or the Governance, Risk Management and Compliance (GRC) Account role, which has a specific role with limited permissions. The User role allows users to view and manage their own account and their quarantine inbox and related settings. Delegation significantly reduces the burden of email administration by safely offloading certain management responsibilities.

Denial of Service (DoS) Attack

An attack designed to bring down a network by flooding it with useless traffic. DDOS (Distributed Denial of Service) is an attack by multiple systems, often compromised by a Trojan, on a single system.

Directory Harvest Attack (DHA)

Scripted attempts designed to steal directory information from unprotected email servers. Spammers send messages with a generic address to your SMTP mail server. For example, spammers will try sending “ed@acme.com,” “sue@acme.com” or other names until the server recognizes a name and accepts a message. Spammers combine dictionaries of common names and tabulate which addresses work and which do not. The technique can yield an overwhelming number of delivery requests, consuming server processing power. Spammers then sell the compiled list of valid email addresses to other spammers, resulting in spam for new and unused accounts.

False Positive

A valid email incorrectly identified as spam. Without any tuning, Email Guard has a false positive rate of 0.01 percent, one of the lowest in the industry.

Hardened Datacenter

Highly secure, fault-tolerant facilities housing IT equipment connected to telecommunications networks. The facilities accommodate web servers, email servers, switches, routers, mass storage and other critical equipment and feature redundancy, backup power, and physical security. Cloud-based infrastructure can combine multiple hardened datacenters in separate geographic regions for high availability and disaster recovery.

Heuristic Rules Analysis

A branch of artificial intelligence whereby expert systems use programs that are self-learning and improve with experience. With spam filtering, the more messages processed the better the results.

HTTPS

The use of Secure Socket Layer (SSL) or Transport Layer Security (TLS) under standard HTTP layering to provide secure web access through encryption.

Image Spam

Image spam, which at one point represented around one third of all Internet traffic, is a technique that hides text, typically a URL, by embedding it within an image. Email Guard uses an Optical Character Reader (OCR) engine to detect this threat. The Update Center also maintains Internet traps known as "honeypots" to build blacklists of image spam senders.

Intent Analysis

Multi-level analysis that recursively follows web links, email addresses and phone numbers embedded in email messages to determine whether they are associated with legitimate entities. Intent analysis also analyzes domain attributes to profile email campaigns.

Key Management

Email encryption requires the sender and recipient to exchange their encryption keys before sending or receiving protected emails. Key management is complex and includes the creation, distribution, safe storage, exchange and use of encryption keys. Email Guard's cloud-based solution for email encryption stores keys centrally and manages them automatically without burdening administrators or users.

LDAP

The Lightweight Directory Access Protocol (LDAP) is an open, industry-standard application protocol for accessing and maintaining distributed directory information services over an IP network. Email Guard's Exchange Accelerator/LDAP Verification feature uses the LDAP support built into Microsoft Exchange to protect against dictionary attacks, which attempt to break into a system by trying every word in a hacker's "dictionary" of possible passwords.

Link Protection

Email Guard's Link Protection, when activated, automatically rewrites any URL embedded within an inbound email and directs it to a safe URL before delivering the email. If the recipient clicks on the embedded URL, Email Guard evaluates the link for validity and reputation. If the domain appears to be valid, the user is directed to that website. If the URL is suspicious, the user is directed to a warning page which shows the actual link and explains why the URL was blocked.

OCR Engine

Embedding text in images is an established technique used by spammers to avoid detection by the text processing mechanisms in anti-spam engines. An optical character recognition (OCR) engine allows a spam filter to analyze the text rendered inside an image. Before submitting images to the OCR engine, Email Guard uses a variety of image processing technologies to normalize images whose text has been obfuscated with speckling, shading or color manipulation. The Update Center also uses advanced image processing algorithms to create fingerprints that can be used to block messages containing image spam.

Phishing

A scam that usually involves a fraudulent email purporting to come from a legitimate sender such as a bank and that either contains links to a spoofed website or else tries to talk the recipient into disclosing private information such as a password or social security number. Email Guard detects Phishing scams by using a specialized Bayesian database. Attacks that include information that suggests familiarity or friendship with the recipient are known as Spear Phishing and account for about 91% of email attacks.

Polymorphic Viruses

A virus that self encrypts by changing its signature to avoid detection by simple virus scanners. Email Guard relies on a supercomputing grid that uses a partial checksum approach to identify the unchanged signature portions of a virus body.

Predictive Sender Profiling

Identifying suspicious behaviors from sources who currently have clean reputations, for example newly infected bots on otherwise legitimate computers.

Quarantining

A quarantined email can be sent to an individual quarantine inbox assigned to the recipient or to a global quarantine mailbox designated by the administrator. Individuals can have quarantine accounts on the Email Guard security gateway, which takes the load off the email server, or on their regular mailbox. In this case, Email Guard alters the subject text of quarantined messages to show that the message could be spam.

Rate Control

The ability to pace inbound or outbound email from a single IP address or email address based on pre-configured frequency thresholds. Delivery of messages from sources that exceed the threshold is delayed in order to stabilize system performance.

Reputation Analysis

The use of databases that include the IP of addresses for known spammers, safe email senders, and web addresses known to be associated with spam. Email from IP addresses with poor reputations or whose content includes web addresses with poor reputations can be identified and handled according to spam policies.

Scoring Thresholds

After an email passes through the Email Guard security gateway it is given a score ranging from 1 to 10 representing the likelihood that the message is spam. The administrator can use scoring thresholds to automate how messages are managed by choosing to allow, tag, quarantine or block emails depending on their score.

Security Perimeter

The outermost edge, such as a firewall or security gateway, over which the enterprise has control over an email message before the message is received from or passed to the Internet.

Signatures

Also called a fingerprint, a series of bytes within a virus that uniquely identifies it and that can be searched for by anti-virus filters.

Spam

Unsolicited and unwanted email usually sent in bulk as part of marketing campaigns but sometimes sent with malicious intent.

Spoofing

A phishing technique whereby an attacker uses an IP address that appears to come from a trusted host. One of the most notorious examples is the exploit from service@aol.com. Spoofing attacks often include links to a fraudulent, or spoofed, web site that impersonates a familiar web site.

Spooling

A disaster recovery feature whereby incoming email is stored temporarily in the cloud or redirected in the event that your on-premises data center is disrupted. Email Guard's Cloud Protection Layer can store all your inbound email for up to four days or direct it to an alternative destination if your email server becomes unavailable. Meanwhile, Progent can provide expert online or onsite support to bring your local system back into service quickly.

Spyware

Software that collects and possibly shares information about a person without specified consent. Common forms of spyware include cookies, adware and Trojans. The collected information is typically used for marketing, but can also be used more maliciously. Major corporations (e.g., Facebook) use spyware routinely and legally. Email Guard includes technology that provides real-time protection from malicious spyware attacks.

SMTP

Short for Simple Mail Transfer Protocol, a protocol for sending email messages between servers. SMTP email servers are a common target of Directory Harvest Attacks.

Tagging

A technique for helping users to identify filtered email that meets some criteria for being spam. Tagged emails are delivered with a customized label appended to the subject field, e.g., [BULK]. This allows users to make their own determination about how to deal with them.

TLS

The Transport Layer Security (TLS) encryption protocol allows secure communication over the Internet. The Email Guard security gateway supports SMTP over TLS to encrypt email traffic between sites over the Internet.

Viruses

Computer code that can replicate itself. Viruses are commonly spread via email and allow an attacker to steal or destroy data or take control of an infected system. Types of viruses include Trojans and worms.

Zero-hour Protection

The ability to develop and distribute defenses against newly discovered spam and viruses within minutes rather than days. The hybrid infrastructure of Email Guard is designed to provide this capability.

Also See:
















© 2002- 2017 Progent Corporation. All rights reserved.

More topics of interest: