Window Sizing for Zstandard Content Encoding

Deployments of Zstandard, or "zstd", can use different window sizes to limit memory usage during compression and decompression. Some browsers and user agents limit window sizes to mitigate memory usage concerns, causing interoperability issues. This document updates the window size limit in RFC8878 from a recommendation to a requirement in HTTP contexts.¶

This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.¶

Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.¶

Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as “work in progress”.¶

This Internet-Draft will expire on February 24, 2025.¶

Copyright (c) 2024 IETF Trust and the persons identified as the document authors. All rights reserved.¶

This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License.¶

Zstandard, or "zstd", specified in [RFC8878], is a lossless data compression mechanism similar to gzip. When used with HTTP, the "zstd" content coding token signals to the decoder that the content is Zstandard-compressed.¶

An important property of Zstandard-compressed content is its Window_Size ([RFC8878], Section 3.1.1.1.2), which describes the maximum distance for back-references and therefore how much of the content must be kept in memory during decompression.¶

The minimum Window_Size is 1 KB. The maximum Window_Size is (1<<41) + 7*(1<<38) bytes, which is 3.75 TB. Larger Window_Size values tend to improve the compression ratio, but at the cost of increased memory usage.¶

To protect against unreasonable memory usage, some browsers and user agents limit the maximum Window_Size they will handle. This causes failures to decode responses when the content is compressed with a larger Window_Size than the recipient allows, leading to decreased interoperability.¶

[RFC8878], Section 3.1.1.1.2 recommends that decoders support a Window_Size of up to 8 MB, and that encoders not generate frames using a Window_Size larger than 8 MB. However, it imposes no requirements.¶

This document updates [RFC8878] to enforce Window_Size limits on the encoder and decoder for the "zstd" HTTP content coding.¶

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here.¶

To ensure interoperability, when using the "zstd" content coding, decoders MUST support a Window_Size of up to and including 8 MB, and encoders MUST NOT generate frames requiring a Window_Size larger than 8 MB (see Section 5.1).¶

This document introduces no new security considerations beyond those discussed in [RFC8878].¶

Note that decoders still need to take into account that they can receive oversized frames that do not follow the window size limit specified in this document and fail decoding when such invalid frames are received.¶

Zstandard was developed by Yann Collet.¶

The authors would like to thank Yann Collet, Klaus Post, Adam Rice, and members of the Web Performance Working Group in the W3C for collaborating on the window size issue and helping to formulate a solution. Also, thank you to Nick Terrell for providing feedback that went into RFC 8478 and RFC 8878.¶